Blog

Analysis of oceanic suspended particulate matter in the western North Pacific using the complex amplitude sensor | Scientific Reports

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Scientific Reports volume  14, Article number: 20055 (2024 ) Cite this article amplitude type

Oceanic suspended particulate matter (SPM) plays important roles in the coupling of climate and biogeochemical cycles via ocean–atmosphere interactions. However, methods for quantifying the properties of SPM in seawater have not yet been well established. Here we present the application of the recently developed complex amplitude sensor (CAS) for analyzing the complex forward-scattering amplitude of individual SPM (0.2–5.0 µm in diameter) obtained at depths of 0–100 m during a research cruise in the western North Pacific. The measured distribution of the complex amplitude indicated that the CAS-derived SPM data could be roughly classified into five major types. Comparison with reference sample’s complex amplitude data and scanning electron microscopy analysis suggested that these types could be attributed mainly to diatom fragments, carbonaceous materials (likely organic matter), mineral dusts, iron oxides, or black carbon. Depth profiles revealed that relatively high concentrations of SPM, presumably dominated by diatom fragments and carbonaceous materials with peak diameters of 0.7–1.0 µm, were typically associated with elevated turbidities and chlorophyll a concentrations. Based on this case study, we discuss the practical advantages and limitations of using the CAS to measure size-resolved concentrations of SPM in seawater and to characterize its composition.

Suspended particulate matter (SPM) is ubiquitous in the ocean and plays a significant role in the biogeochemical cycling of elements in marine environments. The sources of SPM in the ocean can be categorized as either biogenic generated through marine biological activity (e.g., production of SPM associated with phytoplankton, such as diatoms, coccolithophores, dinoflagellates, and cyanobacteria) or external, such as aerosols deposited from the atmosphere (e.g., aerosols released from terrestrial sources) and input from river runoff1,2. Following the biogenic formation or external input, SPM in the ocean undergoes subsequent decomposition or sinking. The decomposition of SPM is an important source of nutrients in the surface ocean, while the sinking of SPM into the ocean interior is a crucial process for carbon removal from the atmosphere2. The direct emission of marine SPM into the atmosphere is thought to be an important source of atmospheric aerosols3. Such marine aerosols can act as cloud condensation nuclei, and some of them may also serve as efficient ice nucleating particles4. Thus, SPM plays important roles in the coupling of climate and biogeochemical cycles via ocean–atmosphere interactions.

Measurements of SPM concentrations and characteristics are an important part of investigations of its roles in climate and biogeochemical cycles. Measuring turbidity is a common method to monitor SPM abundance5,6. Turbidity is quantified based on the amount of light scattered by particles in water, and turbidity is thus a proxy of the bulk mass of SPM7. In marine observations, vertical profiles of turbidity in the water column are often measured using optical backscatter sensors mounted on a conductivity–temperature–depth (CTD) system8. Another indicator of the amount of SPM is the concentration of chlorophyll a (Chl-a), which is a good proxy of the abundance of phytoplankton since phytoplankton species contain plant pigments and Chl-a is the primary photosynthetic pigment present in all forms of algae. Increase of the concentration of Chl-a in surface water is often attributed to bloom-forming phytoplankton such as diatoms and coccolithophores9,10. Measurements of the concentrations of specific elements or substances are also commonly used to characterize SPM. For example, particulate organic carbon, nitrogen, and phosphorus collected on a filter are oxidized and their concentrations measured with an infrared anlyzer11 or a colorimetric technique12. The concentrations of gel-like organic particles such as transparent exopolymer particles can be estimated using staining and light absorption measurements13. The above methods provide a bulk analysis of SPM and do not provide information about the concentration of SPM as a function of particle size.

Information about the particle size-resolved concentrations of SPM is important because particle size is essential for determining the sinking rates of particles in the ocean and atmosphere3,14, and it thus affects vertical and horizontal fluxes. Particles size is also an important determinant of ice nucleating ability15. Techniques that measure individual particles suspended in water have been used to estimate size-resolved concentrations of particles16. For example, the Coulter Counter, which is an electrical instrument that counts particles, is widely used in marine science17,18. Optical methods, which can be used to measure the signal of scattered light from individual particles illuminated by a light source, are also widely used for particle-sizing. The flow cytometer uses a laser beam to illuminate individual particles in a suspension of particles and measures the scattered and fluorescent light to discriminate phytoplankton19,20,21. However, these methods may not be suitable for quantifying both the size-resolved concentrations and specific types of SPM in seawater, particularly non-biological particles.

In this study, we evaluated the applicability of a complex amplitude sensor (CAS22) for particle-sizing and characterization of SPM. The CAS can measure the complex forward-scattering amplitude (complex amplitude S) of single particles suspended in a fluid flow. Yoshida et al.23 have demonstrated that the measured value of S for mineral dust differ from that of black carbon (BC), iron oxides, and bacteria. They have further shown that the size-resolved concentrations of mineral dust can be estimated from S data. Ohata et al.24 have measured the S of atmospheric water-insoluble aerosols; characterized them as either BC-like, mineral-dust-like, or primary biological particles; and quantified the concentration of each kind of aerosol. Moteki et al.25 have identified ambient BC aerosols and have constrained their complex refractive index from S data. Based on these studies, we hypothesized that S measurements could also be used to measure the size-resolved concentrations and composition of SPM in seawater.

In this study, we evaluated the capability of CAS to identify particle composition and quantify size-resolved concentrations of oceanic SPM. For this purpose, we characterized the S data of SPM in seawater and compared them with those of reference materials with known compositions. We also used data from electron microscopic analyses of sampled SPMs to evaluate their composition. Based on these studies, we also show the size-resolved concentrations and depth profiles of SPM. Finally, we discuss the potentials and challenges of using a CAS to measure the concentrations and characteristics of SPM in seawater.

Seawater samples were collected on a research cruise on board R/V Shinsei-maru from 15 July to 2 August 2022 in southeast off Hokkaido, the western North Pacific (Cruise ID: KS-22-10). The seawater in this area is influenced by both the Oyashio Current, a nutrient-rich and cold subarctic ocean current characterized by high biological production, and the Kuroshio Current, a warm ocean current characterized by relatively low biological production. Intensive CTD observations and seawater sampling were conducted at stations C005–C012 (Fig. 1). The temperature–salinity (T–S) diagrams at C005 and C011 were close to that of the Oyashio Current, whereas those at C006, 007, 008, 009, 010, and 012 were close to that of Kuroshio Current (not shown).

Research area and track of cruise. The locations of collecting seawater samples are denoted as black points. This figure was generated by Matplotlib (version 3.8.0, Hunter, J. D., Matplotlib: a 2D graphics environment. Computing in Science and Engineering 9, 2007, https://ieeexplore.ieee.org/document/4160265) and Cartopy (version 0.21.1, MetOffice, UK, https://scitools.org.uk/cartopy/docs/latest/index.html).

Seawater was collected in 2000-mL dark bottles in the following three ways: bucket sampling at the surface (11 samples); a CTD sampling system with Niskin bottles at depths of 10 (9.8–10.5), 20 (18.7–22.8), 30 (29.6–37.7), 50 (48.9–56.1), and 100 (99.1–100.0) m (40 samples); and underway sampling from an intake at a depth of ~ 3 m (25 samples). At station C005, CAS measurements were conducted only for samples collected from depths of 37.7 m or less. Underway sampling was conducted in addition to CTD observations when the vessel was moving. The dark bottles were rinsed twice with seawater prior to sampling. After sampling, we removed large particles with a 12-µm pore size Whatman Nuclepore track-etched membrane filter (47-mm diameter) because such large particles would clog the flow cells (thickness of flow pass = 50 µm) in the CAS instrument. We conducted two types of CAS measurements. One type was “in situ” measurements that were conducted on board immediately (within 4 h) after collecting seawater. Another was “filter” measurements. To perform filter measurements, we first removed large particles from the seawater with the 12-µm pore size filter and concentrated the SPM in the seawater (0.5–1.5 L) on a 0.4-µm pore size Nuclepore track-etched membrane filter (47-mm diameter) and then dried them on a clean bench. The dried samples were transferred to 10-mL sterile centrifuge tubes and stored at ~ 4 °C in the laboratory. These filter samples were immersed in 25 mL of laboratory-purified water (AS ONE Corp.) in a glass vial for measurements with the CAS instrument. We sonicated at frequencies of 28, 45, and 120 kHz for 20 s each (1 min total) to suspend the particles. Finally, the suspension was introduced into the CAS instrument. We also prepared additional filter samples, CAS34 and CAS35, to compare the results of filter measurements made with the CAS, a scanning electron microscope (SEM) coupled with energy-dispersive X-ray spectrometer (EDS), and Coulter Counter. The CAS34 and CAS35 samples were prepared from seawater collected via underway sampling (Fig. 1).

We monitored turbidity with a Seapoint Turbidity Meter (Seapoint Sensors, Inc.) mounted on the CTD frame. Seawater samples for Chl-a analysis were collected from Niskin bottles or (at the surface) a bucket and transferred to 250-mL brown polyethylene bottles. Chl-a samples were gently filtered (volume = 200 mL) by low vacuum pressure (< 0.02 MPa) through a Whatman GF/F filter (25-mm diameter) in a dark room. The filter samples were stored at − 20 °C in the dark. The Chl-a was extracted in 7 mL of N,N-dimethylformamide (DMF) for more than 24 h. After extraction, Chl-a concentrations were measured with a Turner fluorometer (10-AU, Turner Designs), which had previously been calibrated against a pure Chl-a standard (Sigma-Aldrich Co., LLC). We used the fluorometric non-acidification method26 to estimate Chl-a concentrations. Vertical profiles of turbidity and Chl-a concentrations were compared with those of SPM concentrations measured by the CAS instrument.

We prepared some reference samples for comparison of S data. Specifically, we used reference mineral dust samples including NX illite (illite-rich mineral dust27), quartz (SiO2-rich mineral dust, JIS Z 8901 Class3, Association of Powder Process Industry and Engineering), and calcium carbonate (CaCO3, Cat. No. CAH28PB, Kojundo Chemical Laboratory Co., Ltd.). We used iron oxide samples including hematite (Fe2O3, Lot No. 180611, Toda Kogyo Corp.) and goethite (FeOOH, Lot No. 180611, Toda Kogyo Corp.). We used Snomax powder, which is made from the bacterium Pseudomonas syringae28, as a reference for bacteria. We used NIES4227, a strain of Thalassiosira nordenskioeldii Cleve provided by the National Institute for Environmental Studies (NIES) collection through the National BioResource Project (NBRP) of the Ministry of Education, Culture, Sports, Science and Technology, Japan, to obtain reference data for a diatom. This strain was isolated from the Oyashio Current in the western North Pacific29. It is possible that this culture contained certain amounts of contaminants (e.g., culture medium) other than the cultured strain of Thalassiosira. We also obtained S data for ambient BC aerosols that we collected onboard the same research cruise and measured with the CAS instrument25 for identifying BC particles deposited from the atmosphere to the sea surface.

The CAS measures the complex forward-scattering amplitude of single particles suspended in a fluid flow at a wavelength of 632.8 nm. The forward complex amplitude \(S\left(\theta =0^\circ \right)=\text{Re}S+\text{Im}S=\left|S\right|{e}^{i\Delta }\) represents the amplitude and phase shift of the scattered field observed at a forward angle relative to the incident plane-wave field. The S parameter reflects the physical properties of particles such as their complex refractive index, volume, shape, and orientation22,30.

The optical system of the CAS used in this study was identical to that used by Moteki et al.25 The CAS consists of two channels, namely the s-channel for measuring sub-micron particles (0.2–1.0 µm) and the l-channel for super-micron particles (1.0–5.0 µm). Each channel consists of optical components (e.g., mirrors, lens, and beam expander) and a flow cell. Particles suspended in freshwater or seawater are introduced into the flow cell using a peristaltic pump. The flow rate is monitored by a flow meter (SLI 2000, Sensirion) and maintained at ~ 1.9 mL/min. Individual particles are detected when they pass near the focused area of the laser beam (λ = 632.8 nm). The interference of the light scattered forward by individual particles with the incident light (laser beam) is detected by a four-segment photodiode. By analyzing the four interference signals, the complex amplitude S of individual particles is determined. To estimate the size of the beam waist spot size, which needs to be calibrated to determine the S value from the signals, we used the method described by Moteki22 and measured polystyrene latex standard particles with sizes of 0.303, 0.510, 0.803, 1.036, 2.020, 2.994, and 5.010 µm (PSL, Thermo Scientific, 3000 and 4000 series). The resulting beam waist values for the s- and l-channels were estimated to be 3.13 µm and 13.9 µm, respectively. Before each CAS measurement, we introduced Milli-Q water or pure water into the CAS instrument to clean the flow tubes and flow cells. The sample measurement was initiated after the particle detection rate had decreased to less than 0.1 s−1.

The number concentration of SPM detected by the CAS instrument in fluid water (C [m−3]) is expressed by the following equation24:

where R (s−1) is the particle detection rate after correction by the duty cycle of the computer of the CAS instrument for recording particle signal data, A (m2) is the effective detection area perpendicular to the sample flow in the CAS flow cell, and U (ms−1) is the average velocity of the sample water in the flow cell. The effective detection area A depends on the alignment of optical systems and thresholds set in the software to detect waveform signals of particles passing near the center of the laser beam. The thresholds were identical to those set by Moteki25. We experimentally determined the value of A by measuring R for suspensions of various sizes of PSLs. The number concentrations of which, C, were determined based on the Lambert–Beer law31. We used PSLs with sizes of 0.303, 0.510, 0.803, 1.036, 2.020, 2.994, and 5.010 µm. Supplementary Fig. 1 shows the experimentally determined values of A in the s-channel (As-ch) and the l-channel (Al-ch). The As-ch for 0.303-µm PSLs was 2.82 × 10−11 m2, and the mean value for 0.510-, 0.803-, and 1.036-µm PSLs was 4.52 × 10−11 m2. The mean value of Al-ch for 0.803- and 1.036-µm PSLs was 1.50 × 10−10 m2, and that of 2.020-, 2.994-, and 5.010-µm PSLs was 2.38 × 10−10 m2. In both channels, the A values were low for the smaller PSL. A low signal-to-noise (S/N) ratio of waveform signals for such small particles could lead to failure to detect the particle signal and decrease the A value24. The size dependence of A is likely to vary as a function of particle composition because the relationship between the S/N ratio of the signal and particle size varies among particle compositions. For this reason, we used As-ch = 4.52 × 10−11 m2 and Al-ch = 2.38 × 10−10 m2 and did not consider the lower A value determined by the smaller particles when measuring the SPM. The derived number concentration for small SPMs (< ~ 0.5 µm for s-channel and < ~ 1.0 µm for l-channel) was thus likely underestimated.

To validate the procedure for determining the number concentration, we conducted simultaneous measurements of PSL suspensions by CAS and Coulter Counter (Multisizer 4, Beckman Coulter, Inc.) in the laboratory. The aperture size of the Coulter Counter was 30 µm, and the detectable particle size range was 0.6–24 µm. We prepared monodispersed suspensions containing 1.036-µm PSLs and polydispersed suspensions containing 0.803-, 1.036-, and 2.994-µm PSLs. We diluted these suspensions with ISOTON II electrolyte to measure electrical impedance with a Coulter Counter. Because the PSL particles were spherical and their complex refractive index m had been measured, it was straightforward to determine their particle size from measured S data points. We first calculated the theoretical S curve, which is a series of points of (ReS, ImS) with diameter D as parameter, based on Mie theory, assuming the complex refractive index of a PSL in vacuum at λ = 632.8 nm to be 1.5854 + 6.1764 × 10−7i based on experimental data32. We then projected each measured data point on the S curve to determine the particle size. Combining the particle size and particle detection rate allowed us to determine the size-resolved concentration. Supplementary Fig. 2a,b show size-resolved concentrations, dN/dlogD, where D represents the diameter of the particle and dN represents the number concentration of particles within each size bin (dlogD). Both the CAS and Coulter Counter successfully quantified the sizes of the PSLs. We further derived the number concentration N for PSLs of D = 0.803, 1.036, and 2.994 µm by integrating dN/dlogD in the D ranges of 0.6–0.86, 0.86–1.2, and 2.6–3.5 µm, respectively. Supplementary Fig. 2c shows the scatter plot of the N derived by the CAS and Coulter Counter. For the l-channel, we also derived a modified N for PSLs of D = 0.803 and 1.036 µm by using an Al-ch value of 1.50 × 10−10 m2 instead of 2.38 × 10−10 m2. Use of the modified N caused the data points to lie close to the 1:1 line in the scatter plot. Combining the data points for the s-channel and the modified l-channel led to a linear-fit slope of 0.8727 and an extremely high R2 value of 0.996. We obtained good agreement between the size-resolved concentrations of PSL determined by CAS and Coulter Counter.

The complex amplitude of S reflects the physical properties of particles, including their complex refractive index, size, and shape. It is expected that a group of particles with a specific composition will form a characteristic S cluster on the complex S plane23,24,25,30,33. Because there is some arbitrariness in how observed data are divided into clusters, we set boundaries on the complex S plane to classify the clusters for the SPM. To determine the particle size for each S data point, we used a forward model that calculates the S value from the assumed complex refractive index and shape. Moteki et al.25 estimated the complex refractive index of ambient BC aerosols measured by CAS onboard the same research cruise. They used a curve fit to the S data and then conducted Bayesian inference with the SPHPACK shape model to determine the most plausible parameter vector (size, complex refractive index, and shape) for the S data points on the fitted curve. We used the relation between the S value and its plausible particle size expressed as the volume equivalent diameter inferred by Moteki et al.25 based on the data measured on 27 July 2022 (Fig. 7a from Moteki et al.25) to determine the particle size of the BC in the SPM in seawater. On the other hand, detailed forward models that fully consider the particle’s physical properties (volume, shape, and complex refractive index) have not yet been developed for non-BC SPMs. We estimated the size of non-BC particles by assuming a spherical shape and zero imaginary part of the complex refractive index (i.e., a non-absorbing sphere). In this approach, we prepared a table of theoretical S values for non-absorbing spheres with diameters in the range 0.1 µm < D < 10.0 µm and refractive indices in the range 1.34 < m < 2.40 using the Mie theory, where m is the complex refractive index of particles in vacuum. We then determined the most likely refractive index mml and diameter D of each measured particle by minimizing the distance between the measured and theoretical S values. We note that S depends on the ratio of the complex refractive index of particles in vacuum m to that of the medium of the suspension mmed. In calculating the theoretical S values, we assumed mmed to be the refractive index of pure water for the filter measurements and seawater for the in situ measurements. We used mmed = 1.33154 for pure water at λ = 632.8 nm34 and mmed = 1.3373 for seawater at λ = 640 nm, temperature = 20 °C, and salinity of 34.0‰35. For each cluster, a whole size-resolved number concentration was constructed by combining the results from the s- and l-channels. The detectable ranges of S partly overlapped between the s- and l-channel configurations. In this study, data points detected in the s-channel were used in the range 0 µm < ReS < 1.5 µm and 0 µm < ImS < 0.5 µm, and those detected in the l-channel were used in the range 1.5 µm < ReS < 10.0 µm and 0.5 µm < ImS < 5.0 µm. The reason for setting the upper limits for the l-channel was that large particles that exceeded the upper limit did not satisfy the validity criteria of the plane wave approximation when the S value was determined from the signal waveform22. All computational analyses of S data for characterizing particle composition and quantifying size-resolved concentration were carried out using the Python programming language (version 3.10.12, https://www.python.org).

To identify the particle compositions corresponding to the S clusters, we analyzed CAS34 and CAS35 samples using a SEM (SU 3500, Hitachi High-Technologies Co.) coupled with an EDS (X-Max50, Horiba, Ltd.). Individual particles over the filters having contrast above the threshold intensity in the SEM images were automatically detected, and their area equivalent diameter (> 0.55 µm) and elemental spectra were measured using multipoint analysis software (Oxford INCA software). Based on the elemental spectra, we classified particles into specific types and determined each size distribution.

Figure 2a–d shows the S data points for both in situ and filter measurements with all SPM samples collected from this study. The distribution of S data points was expressed as normalized dN/(dReSdImS), where dN (mL−1) was the number of particles within each grid (dReSdImS) on the complex S plane. When characterizing the data points for the s-channel, we excluded the data points in the boundary defined by the elliptic function (x/0.2)2 + (y/0.025)2 = 1, where x = ReS and y = ImS, because the data points near the origin of the complex plane were too concentrated to classify particle Types. We categorized the other data points into Types 0–4 as follows. We classified the data points located to the right of the theoretical S curve for the sphere with m = 1.38, between m = 1.38 and 1.43, and between m = 1.43 and 1.60 as Types 0, 1, and 2, respectively, as denoted in Fig. 2. These theoretical curves were calculated based on Mie theory. In the s-channel, we further confirmed some data points distributed to the left of m = 1.60. We classified those data points as Type 3 if they were located to the right of the lines y = − 5.5x + 0.6 or y = 0.72x − 0.0072, where x = ReS and y = ImS. We also classified the data points as Type 4 if they were located to the left of both of these lines. We did not define Types 3 and 4 for the l-channel because of the limited number of data points. Among all samples measured by in situ methods, the 25th–75th percentiles of the number concentrations of Types 0, 1, 2, 3, and 4 data points combined in both the s- and l-channels were 2.4 × 104–6.7 × 104, 4.8 × 104–9.7 × 104, 6.0 × 103–1.5 × 104, 3.8 × 102–1.2 × 103, 1.5 × 102–4.1 × 102 mL−1, respectively. For the filter measurements, the corresponding values were 1.3 × 104–3.7 × 104, 1.5 × 104–4.8 × 104, 4.7 × 103–1.4 × 104, 9.3 × 102–5.3 × 103, 2.7 × 101–1.8 × 102 mL−1, respectively.

(a–d) Distribution of S data points combined for (a,b) all in situ samples and (c,d) all filter samples. Boundaries for classifying S data points are represented as dashed curves and lines. The boundary curves are theoretical S curves for sphere particles with complex refractive index of m. (e,f) Median of ReS values as functions of ImS values for reference samples. The black rectangle boxes at lower left in l-channel panels show the outer boundaries of s-channel panels.

Figure 2e,f present the S for the reference samples, which are depicted as median ReS values as functions of ImS. The S data points for Snomax were concentrated near the boundary between Types 0 and 1 (i.e., around m = 1.38) in the s-channel, and they were predominantly distributed in the Type 0 region in the l-channel. The data points for NX illite were distributed near the boundary between Types 1 and 2 in both the s- and l-channels (i.e., around m = 1.43). Most data points for the calcium carbonate and quartz samples were distributed in Type 2 regions in both the s- and l- channels. Most data points of the hematite and goethite were distributed in the Type 3 region in the s-channel. Most data points for small BC particles (i.e., ImS < 0.15 µm) were distributed in the Type 4 region, whereas the data for larger BC particles (i.e., ImS > 0.15 µm) were distributed in the Type 3 region. The distinct distributions of the smaller BC and iron oxide made it necessary to separate Types 3 and 4. Although the data points for diatom culture strain NIES4227 exhibited a scattered distribution (Supplementary Fig. 3), 56% of the data points were classified as Type 0, whereas 19%, 19%, 4.6%, and 0.25% of the data points were classified as Types 1, 2, 3, 4, respectively, when the data from the s- and l-channels were combined.

Figure 3a presents representative SEM images of the analyzed particles for CAS34 and CAS35. Representative EDS spectra are shown in Supplementary Fig. 4. Because the EDS spectra contained background signals attributable to carbon (C) and oxygen (O) from the polycarbonate filter and aluminum (Al) from the stage holder, it was difficult to accurately quantify the signals of C, O, and Al from the particles. Based on these spectra, we categorized the detected particles into several types. We classified more than half of the particles (64% for CAS34 and 57% for CAS35) as “carbonaceous.” Weak signals from nitrogen (N) tended to be commonly present in the carbonaceous particles (Supplementary Fig. 4) and signals from phosphorous (P) were also detected in some of them. Because phosphorus and nitrogen are elements that make up biological organic compounds36, we hypothesized that these carbonaceous particles were mainly attributable to biological organic compounds. We classified particles from which there were strong signals of silicon (Si) as “Si-rich” particles, and they accounted for 26% (CAS34) and 29% (CAS35) of the total analyzed particles. SEM images of Si-rich particles (Fig. 3a) indicated that they might be composed of mineral dust or diatoms. Mineral dust particles typically exhibit angular shapes with a brighter appearance in back-scattered electron images because of their high density. Some Si-rich particles in the super-micron size range had shapes characteristic of diatoms (Fig. 3a right). However, we did not distinguish between mineral dust and diatoms because it was difficult for our SEM–EDS analysis to unambiguously distinguish between small (< ~ 1 µm) silicate mineral dust and a fragment of diatom based on their shapes and compositions. Particles from which we detected strong signals from other elements such as sodium, potassium, and chlorine were classified as “Others” and accounted for 11% (CAS34) and 13% (CAS35) of all analyzed particles.

SEM–EDS analyses of representative particles found in CAS34 and CAS35 samples. (a) Representative SEM images of carbonaceous and Si-rich (mineral dust and diatom) particles (indicated by arrows). Red arrows indicate particles whose spectra are shown in Supplementary Fig. 4. (b) Comparison of size-resolved concentrations of the particles determined by CAS and SEM–EDS.

We compared the size-resolved number concentration of each Type derived from CAS based on the boundaries defined above and SEM–EDS (Fig. 3b). We determined the CAS-derived sizes of Type 4 particles by assuming that they were BC (see the next section) and used the relationship between the S value and volume-equivalent diameter inferred by Moteki et al.25 For the sizes of Types 0–3, we assumed that these particles were non-absorbing spheres. We acknowledge a significant uncertainty in estimating the particle size of Type 3 particles (likely light-absorbing iron oxides, see the next section) on the assumption that they are non-absorbing spheres, and thus more accurate estimation of the particle size would require a suitable model that simulates the S for realistic iron oxide particles. Nevertheless, the combined size-resolved concentrations of Types 3 and 4 (Type 3 + Type 4) were so low that they were difficult to discern in the plots. We compared the CAS results with the size-resolved concentrations of carbonaceous and Si-rich types determined by SEM–EDS analysis. Note that the size-resolved concentrations derived from the SEM–EDS analysis provide only the relative concentrations of SPM in the seawater. For both CAS34 and CAS35 samples in the size range 0.55–5.0 µm, the Type 1 (carbonaceous) particles were approximately 2.5-fold and 1.9-fold more abundant than the Type 0 + Type 2 (Si-rich) particles, respectively. The size distributions of the Type 1 (carbonaceous) particles exhibited a steep decrease from 0.55 to 1.0 µm, whereas the Type 0 + Type 2 (Si-rich) particles exhibited a gradual decrease. These distributions of Type 0 + Type 2 and Type 1 were similar to those of Si-rich particles and carbonaceous particles measured using SEM, respectively.

In Supplementary Fig. 5, we compare the absolute number concentrations for CAS34 samples determined by CAS and Coulter Counter. Despite the differences of the measuring principles between the CAS and Coulter Counter, the size-resolved concentrations derived from both instruments were consistent for particles larger than 0.6 µm.

Based on the above results, we propose the Major, Minor, and Possible compositions corresponding to each Type (Table 1). For all Types, the Major composition of the adjacent Type on the complex S plane are included in the Minor composition. A composition, which may be either the Major or Minor composition, but there is no clear evidence from this study, is categorized as the Possible composition.

There are several candidates for the Major composition of Type 0. For example, bacteria are carbonaceous, and the S data points of Snomax were distributed in the regions of both Type 0 and 1. For this reason, we included carbonaceous particles as a Major composition of Type 0 as well as Type 1. In addition, we assumed diatom fragments to be a Major composition of Type 0 because more than 50% of the S data points of the NIES4227 sample were classified as Type 0 (Supplementary Fig. 3). In contrast, ~ 20% of the S data points for NIES4227 were classified as Types 1 and 2. We therefore included diatom fragments in the Minor compositions of both Types 1 and 2 as well. The typical refractive index of opal, a component of diatom shells, lies in the range 1.41–1.4637. Diatom shells also contain numerous internal cavities (e.g., Fig. 3a). The ImS/ReS ratio of particles with such shapes tends to be lower than that of spherical (compact) particles33. The S data points for such particles would lie to the right of the theoretical curve corresponding to m = 1.46. This expectation was consistent with the S data points of NIES4227, which were distributed in Type 0 and 1 regions to the right of the theoretical curve corresponding to m = 1.43. It is unclear why ~ 20% of the data points were distributed in the Type 2 region, which is bounded by m = 1.60 and m = 1.43, but it is possible that particles other than diatoms in the culture sample were contributing to the distribution in this region.

We considered carbonaceous particles to be a Major composition of Type 1. In the analyses of CAS34 and CAS35 samples, the highest number of particles were classified into Type 1 in the CAS measurements and into carbonaceous in the SEM analyses. Moreover, the peaks of the size-resolved concentrations of the Type 1 particles derived from the CAS measurements closely resembled those of the relative size-resolved concentrations of carbonaceous particles derived from the SEM–EDS analyses (Fig. 3b). In marine environments, SPM is composed primarily of organics1,38. Within the size range of particles analyzed in this study (0.2–5.0 µm), the composition of organic particles included colloidal particles, bacteria, cyanobacteria, eukaryotic picoplankton, and detritus21,39. Aas37 has calculated the refractive index of phytoplankton, assuming a typical volume fraction of water (0.6–0.8), to be 1.377–1.417. The close alignment of this range of refractive indices with the lower limit (the theoretical curve with m = 1.38) and upper limit (m = 1.43) of Type 1 on the complex S plane supports our hypothesis that carbonaceous particles are a Major composition of Type 1. Snomax (bacteria), which do not contain chlorophyll, are expected to have a lower refractive index than phytoplankton, and that lower refractive index would probably lead to a lower ImS/ReS ratio, which might explain why data points for Snomax were located in the regions of both Types 0 and 1.

We assumed mineral dust to be a Major composition of Type 2 based on the S data for the NX illite, quartz, and calcium carbonate samples (Fig. 2e,f). Note that the data points for NX illite were distributed on the boundary between Types 1 and 2. The lower number concentration of particles in Type 2 compared to Type 1 in the SPM samples might have obscured the S clusters of mineral dust and led to uncertainty in identification of the composition of Type 2 particles. Although calcium carbonate is a ubiquitous component of mineral dust, it is also a component of coccoliths, which form the shells of coccolithophores. For this reason, we include coccolith as a Possible composition of Type 2. Further measurements of coccolithophore cultures are necessary to confirm the Type associated with coccoliths.

We categorized iron oxides and BC as Major components of Types 3 and 4, respectively, because reference data for iron oxides belonged to Type 3, and data for atmospheric BC smaller than ~ 0.4 µm during the same research cruise25 belonged to Type 4. We note that the amounts of SPM categorized as Type 3 or 4 were much lower than those of other Types and that the data points for large BC (> ~ 0.4 µm) occasionally overwhelmed those distributed in the Type 3 region, resulting in a technical difficulty for the characterization of particle compositions for Types 3 and 4 contained in the seawater samples.

Figure 4 shows size-resolved number concentrations of each type by depth for all in situ measurements. Following the CAS34 and CAS35 analyses, we determined particle sizes for Type 4 as BC and Type 0–3 as non-absorbing spheres. The sizes of Types 0, 1, 2, 3, and 4 were approximately 0.7–5.0, 0.5–3.0, 0.7–2.0, 0.2–0.5, and 0.2 µm, respectively. The peak sizes of Types 0, 1, and 2 were approximately 0.8, 0.7, and 0.5 µm, respectively. Types 0–2 exhibited relatively consistent size-resolved concentration shapes across different depths. Types 3 and 4 showed no distinct peak because of their low concentrations, and they were seldom detected below the surface. Supplementary Fig. 6 shows the size-resolved volume concentrations dV/dlogD obtained by multiplying dN/dlogD by πD3/6. The volume concentrations for Types 3 and 4 were too low to be visible in the figure because the values were more than two orders of magnitude lower than those for Types 0 and 1. For Types 0–2, the peaks observed in the size-resolved number concentrations became less distinct, and significant volume concentrations were even observed near the upper limit of the size-resolved concentrations (~ 5 µm).

Size-resolved number concentrations of Type 0–4 particles obtained by the in situ measurements shown by depth. Solid line and shaded area represent median and 25–75 percentile values, respectively.

We further investigated the difference of size-resolved number concentrations between the filter and in situ measurements for Types 0–2 (Supplementary Fig. 7). The ratio of dN/dlogD determined by the filter and in situ measurements for Types 0–2 increased and exceeded 1.0 as the particle size approached the lower limits of the CAS particle sizing. This feature was particularly evident in Type 1, where the ratio decreased notably as the particle size increased beyond ~ 1.0 µm. This pattern suggested that the filtration and sonication used for filter measurements might have some impact on the number-size distribution of fragile carbonaceous particles38,40. Filter measurements, however, have the advantage of being able to yield concentrated SPM suspensions and more data points than do in-situ measurements (Fig. 2).

Figure 5 shows vertical profiles of SPM number and volume concentrations determined by in situ measurements, and vertical profiles of Chl-a concentrations and turbidity at eight CTD stations (C005–C012). The volume concentration was derived by integrating the dV/dlogD over the whole size range obtained by s- and l-channel measurements. The Chl-a concentrations were lower at station C011 than at other stations because C011 was located in a warm eddy detached from the Kuroshio Extension. The high Chl-a concentration at C005, where the T–S diagram was close to that of Kuroshio Current, might have been due to the intrusion of nutrient-rich Oyashio water. The subsurface chlorophyll maximum layers (SCMLs), which were created mainly by a combination of light attenuation and the vertical gradient of nutrient concentrations, were clearly apparent at depths of 10–60 m. The depths of the SCML were typical for waters ranging from inshore to offshore41. The number and volume concentrations of SPM correlated relatively well with turbidities, indicating that CAS-derived volume concentrations were properly quantified. Furthermore, the peak depths of SPM concentrations were located near the depths of the SCML. At C011, there was a clear peak of the SPM number concentration at 20 m, but the peaks of volume concentration and Chl-a were weak, and there was no apparent peak of turbidity. More sub-micron particles and fewer super-micron particles were observed at a depth of 20 m at C011 than at the other stations (not shown). We could not confirm this abundance of sub-micron particles by the particle volume dependence of the Chl-a concentrations or the particle surface area or volume dependence of the turbidity.

Vertical profiles of number and volume concentrations of SPM obtained by the in situ measurement, comparing with the vertical profiles of turbidity and Chl-a concentration.

Figure 6 shows vertical profiles of SPM volume concentrations for Types 0 and 1. Types 2–4 are not shown because their concentrations were low. At depths shallower than the SCML, the distributions of the concentrations of Type 1 were similar to those of the Chl-a concentrations, although there were not strong peaks of Type 1 at the depths of the SCMLs in some profiles (e.g., C005, C006, C007, and C010). At greater depths, both the Type 1 and Chl-a concentrations decreased sharply. The patterns of the concentrations of Type 0 were similar to those of turbidity. Supplementary Fig. 8 shows the correlations between the Chl-a concentrations and the SPM volume concentrations by Type for all samples measured in this study. The R2 values (p values) for the correlations associated with the linear regressions were 0.48 (1.9 × 10−8), 0.64 (2.6 × 10−12), 0.37 (3.3 × 10−6), 0.02 (0.096), and 0.00 (0.35) for Type 0, 1, 2, 3, and 4, respectively. These values suggested that the origins of some of the Type 0–2 particles may have been phytoplankton, whereas Type 3 and 4 particles were unrelated to phytoplankton. This interpretation is consistent with our hypothesis that the compositions of Types 0–2 included carbonaceous and diatom fragments, whereas those of Types 3 and 4 did not (Table 1).

Vertical profiles of volume concentrations for Types 0 and 1 obtained by the in situ measurements, comparing with the vertical profiles of turbidity and Chl-a concentration. Types 2–4 are not shown due to their low concentrations.

Figure 7 shows the volume concentrations of Types 0–4 at a depth of 0 m (sea surface) and a depth of 10 m. The values for Types 2–4 have been magnified to facilitate visual analysis. At all stations, the volume concentrations of Types 3 and 4 were distinctly higher at the sea surface than at a depth of 10 m. The volume concentrations of Types 0 and 1 showed no such tendency. At the sea surface, the highest concentrations of Types 3 and 4 occurred at C005. The Type 3 and 4 particles observed at C005 were likely deposited from the atmosphere via rainfall for the following three reasons. First, iron oxides and BC, the compositions of Types 3 and 4, are known to be water-insoluble components of aerosols42. Second, a frontal system associated with precipitation passed through the research area on 20 July, the day before the observations at C005. The resulting precipitation might have formed a freshwater lens at the sea surface43, where aerosol particles in the raindrops might have accumulated. Third, the mean (standard deviation) of the 1-min wind speed measured on the vessel from 22:00 on 20 July (JST), after the frontal system had passed, to 10:00 on 21 July (JST), when observations were made at C005, was 2.1 (0.95) m s−1. Such weak winds would have probably not perturbed the near-surface stratification44. During the observations at stations C007 to C012, more than three days had passed since the passage of the last frontal system (23 July). However, we nevertheless observed fluctuations of the concentrations of Type 3 and 4 particles at the sea surface. It is possible that Type 3 and 4 particles remained in the surface layer such as microlayer45 around these stations.

Temporal variations in the volume concentration of each Type of SPM at a depth of 0 m and 10 m. The date are displayed in JST.

Mineral dust is a major component of water-insoluble aerosols, and it is possible that it was deposited on the sea surface through precipitation. The volume concentration of Type 2 particles was higher at station C005 than at C006, similar to the volume concentrations of Types 3 and 4 (Fig. 7). This similarity suggests that some Type 2 particles, likely mineral dust, might also have been deposited from the atmosphere. However, unlike Types 3 and 4, Type 2 particles were not concentrated at the ocean surface. The correlation of Type 2 particle concentration with Chl-a concentration suggests that particles of biological origin, such as carbonaceous particles and coccolithophores, may also be classified as Type 2.

In this study, we evaluated the capability of CAS to characterize the SPM in seawater. We confirmed that the size-resolved concentrations of SPM derived from the CAS measurements were reasonably consistent with the results of Coulter Counter measurements. The number and volume concentrations of SPM derived by CAS were also well correlated with turbidity. These results indicated that CAS could provide reliable quantification of SPM in the ocean. We further classified the CAS-derived SPM into five major types (Type 0 to Type 4). Based on the comparison with the reference sample’s complex amplitude data and SEM–EDS analysis, we propose that these Types could be attributed mainly to diatom fragments, carbonaceous materials (likely organic matter), mineral dusts, iron oxides, or black carbon.

The majority of SPM in the ocean is thought to consists of organic particles1,38,46,47. Our results indicate that particles in Types 0 and 1, which were likely to be dominated by carbonaceous particles and diatom fragments, were prevalent in seawater. The volume concentrations of Type 0 and 1 particles correlated relatively well with the turbidity and Chl-a concentration, which is consistent with our hypothesis that the main compositions of these Types include carbonaceous (likely organic matter). Type 2–4 particles were less abundant compared to Type 0 and 1 particles. Nevertheless, CAS could observe relatively high amounts of Type 2–4 particles at sea surface, which were probably deposited from the atmosphere. Although there are still some technical difficulties with the boundaries for classifying particle Types in the complex S plain and there could be uncertainties in identifying particle compositions, these results suggested that CAS measurements have potential to quantify size-resolved concentrations of SPM from both biogenic sources (carbonaceous and diatom fragments) and external sources (mineral dust, iron oxides, and black carbon) in the size range of 0.2–5.0 µm in diameter. The classification of these compositions is important for ocean biogeochemical studies because each composition has distinctive features. For example, organic particles play important role in carbon cycle2, diatoms affect silicon and carbon cycles through biological pumps48, mineral dust and iron oxides have potential to supply micronutrient to phytoplankton49,50, and BC can adsorb organic particles and alert their size45. Further development of CAS measurements is expected to be useful for observing the abundance and compositions of SPMs in the ocean. In addition, further measurements of SPM in various sea areas worldwide are needed to assess whether the CAS measurements are universally applicable.

The size-resolved concentrations of SPM used in this study are available at https://www.jamstec.go.jp/datadoi/doi/https://doi.org/10.17596/0003381.html. Raw data of individual particles are available from the corresponding author.

Iwamoto, Y. & Uematsu, M. Spatial variation of biogenic and crustal elements in suspended particulate matter from surface waters of the North Pacific and its marginal seas. Prog. Oceanogr. 126, 211–223 (2014).

Sigman, D. & Hain, M. The biological productivity of the ocean. Nat. Educ. Knowl. 3, 21 (2012).

Alsante, A. N., Thornton, D. C. O. & Brooks, S. D. Ocean aerobiology. Front. Microbiol. 12, 764178 (2021).

Article  PubMed  PubMed Central  Google Scholar 

Brooks, S. D. & Thornton, D. C. O. Marine aerosols and clouds. Annu. Rev. Mar. Sci. 10, 289–313 (2018).

Liu, Q., Kandasamy, S., Lin, B., Wang, H. & Chen, C.-T.A. Biogeochemical characteristics of suspended particulate matter in deep chlorophyll maximum layers in the southern East China Sea. Biogeosciences 15, 2091–2109 (2018).

Article  ADS  CAS  Google Scholar 

Fettweis, M. et al. Uncertainties associated with in situ high-frequency long-term observations of suspended particulate matter concentration using optical and acoustic sensors. Prog. Oceanogr. 178, 102162 (2019).

Jafar-Sidik, M., Gohin, F., Bowers, D., Howarth, J. & Hull, T. The relationship between Suspended Particulate Matter and Turbidity at a mooring station in a coastal environment: Consequences for satellite-derived products. Oceanologia 59, 365–378 (2017).

Haalboom, S. et al. Suspended particulate matter in a submarine canyon (Whittard Canyon, Bay of Biscay, NE Atlantic Ocean): Assessment of commonly used instruments to record turbidity. Mar. Geol. 434, 106439 (2021).

Brown, C. W. & Yoder, J. A. Coccolithophorid blooms in the global ocean. J. Geophys. Res. Oceans 99, 7467–7482 (1994).

Article  ADS  CAS  Google Scholar 

Hallegraeff, G. M. Seasonal study of phytoplankton pigments and species at a coastal station off Sydney: Importance of diatoms and the nanoplankton. Mar. Biol. 61, 107–118 (1981).

Menzel, D. W. & Vaccaro, R. F. The measurement of dissolved organic and particulate carbon in seawater1. Limnol. Oceanogr. 9, 138–142 (1964).

Pujo-Pay, M. et al. Improvement of the wet-oxidation procedure for simultaneous determination of particulate organic nitrogen and phosphorus collected on filters. Mar. Ecol. Prog. Ser. 105, 203–203 (1994).

Article  ADS  CAS  Google Scholar 

Passow, U. & Alldredge, A. L. A dye-binding assay for the spectrophotometric measurement of transparent exopolymer particles (TEP). Limnol. Oceanogr. 40, 1326–1335 (1995).

Article  ADS  CAS  Google Scholar 

Omand, M. M., Govindarajan, R., He, J. & Mahadevan, A. Sinking flux of particulate organic matter in the oceans: Sensitivity to particle characteristics. Sci. Rep. 10, 5582 (2020).

Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

Murray, B. J., O’Sullivan, D., Atkinson, J. D. & Webb, M. E. Ice nucleation by particles immersed in supercooled cloud droplets. Chem. Soc. Rev. 41, 6519–6554 (2012).

Article  CAS  PubMed  Google Scholar 

Reynolds, R. A., Stramski, D., Wright, V. M. & Woźniak, S. B. Measurements and characterization of particle size distributions in coastal waters. J. Geophys. Res. https://doi.org/10.1029/2009JC005930 (2010).

Sheldon, R. W., Prakash, A. & Sutcliffe, W. H. Jr. The size distribution of particles in the ocean1. Limnol. Oceanogr. 17, 327–340 (1972).

Beardsley, G. F. Jr., Pak, H., Carder, K. & Lundgren, B. Light scattering and suspended particles in the eastern equatorial Pacific Ocean. J. Geophys. Res. 75, 2837–2845 (1970).

Smyth, T. J., Tarran, G. A. & Sathyendranath, S. Marine picoplankton size distribution and optical property contrasts throughout the Atlantic Ocean revealed using flow cytometry. Appl. Opt. 58, 8802–8815 (2019).

Article  ADS  CAS  PubMed  Google Scholar 

Swalwell, J. E., Ribalet, F. & Armbrust, E. V. SeaFlow: A novel underway flow-cytometer for continuous observations of phytoplankton in the ocean. Limnol. Oceanogr. Methods 9, 466–477 (2011).

Veldhuis, M. J. W. & Kraay, G. W. Application of flow cytometry in marine phytoplankton research: Current applications and future perspectives. Sci. Mar. 64, 121–134 (2000).

Moteki, N. Measuring the complex forward-scattering amplitude of single particles by self-reference interferometry: CAS-v1 protocol. Opt. Express 29, 20688–20714 (2021).

Article  ADS  CAS  PubMed  Google Scholar 

Yoshida, A., Moteki, N. & Adachi, K. Identification and particle sizing of submicron mineral dust by using complex forward-scattering amplitude data. Aerosol Sci. Technol. https://doi.org/10.1080/02786826.2022.2057839 (2022).

Ohata, S. et al. Evaluation of a method to quantify the number concentrations of submicron water-insoluble aerosol particles based on filter sampling and complex forward-scattering amplitude measurements. Aerosol Sci. Technol. https://doi.org/10.1080/02786826.2023.2223387 (2023).

Moteki, N., Ohata, S., Yoshida, A. & Adachi, K. Constraining the complex refractive index of black carbon particles using the complex forward-scattering amplitude. Aerosol Sci. Technol. https://doi.org/10.1080/02786826.2023.2202243 (2023).

Welschmeyer, N. A. Fluorometric analysis of chlorophyll a in the presence of chlorophyll b and pheopigments. Limnol. Oceanogr. 39, 1985–1992 (1994).

Article  ADS  CAS  Google Scholar 

Broadley, S. L. et al. Immersion mode heterogeneous ice nucleation by an illite rich powder representative of atmospheric mineral dust. Atmos. Chem. Phys. 12, 287–307 (2012).

Article  ADS  CAS  Google Scholar 

Wex, H. et al. Intercomparing different devices for the investigation of ice nucleating particles using Snomax® as test substance. Atmos. Chem. Phys. 15, 1463–1485 (2015).

Article  ADS  CAS  Google Scholar 

Suzuki, S. et al. Depth-dependent transcriptomic response of diatoms during spring bloom in the western subarctic Pacific Ocean. Sci. Rep. 9, 14559 (2019).

Article  ADS  PubMed  PubMed Central  Google Scholar 

Potenza, M. A. C., Sanvito, T. & Pullia, A. Measuring the complex field scattered by single submicron particles. AIP Adv. 5, 117222 (2015).

Ohata, S., Moteki, N., Schwarz, J., Fahey, D. & Kondo, Y. Evaluation of a method to measure black carbon particles suspended in rainwater and snow samples. Aerosol Sci. Technol. 47, 1073–1082 (2013).

Article  ADS  CAS  Google Scholar 

Zhang, X., Qiu, J., Li, X., Zhao, J. & Liu, L. Complex refractive indices measurements of polymers in visible and near-infrared bands. Appl. Opt. 59, 2337–2344 (2020).

Article  ADS  CAS  PubMed  Google Scholar 

Moteki, N. Capabilities and limitations of the single-particle extinction and scattering method for estimating the complex refractive index and size-distribution of spherical and non-spherical submicron particles. J. Quant. Spectrosc. Radiat. Transf. 243, 106811 (2020).

Hale, G. M. & Querry, M. R. Optical constants of water in the 200-nm to 200-microm wavelength region. Appl. Opt. 12, 555–563 (1973).

Article  ADS  CAS  PubMed  Google Scholar 

Austin, R. W. & Halikas, G. The Index of Refraction of Seawater (Scripps Institution of Oceanography, 1976).

Paytan, A. & McLaughlin, K. The oceanic phosphorus cycle. Chem. Rev. 107, 563–576 (2007).

Article  CAS  PubMed  Google Scholar 

Aas, E. Refractive index of phytoplankton derived from its metabolite composition. J. Plankton Res. 18, 2223–2249 (1996).

Koike, I., Hara, S., Terauchi, K. & Kogure, K. Role of sub-micrometre particles in the ocean. Nature 345, 242–244 (1990).

Stramski, D., Boss, E., Bogucki, D. & Voss, K. J. The role of seawater constituents in light backscattering in the ocean. Prog. Oceanogr. 61, 27–56 (2004).

Yamasaki, A. et al. Submicrometer particles in northwest Pacific coastal environments: Abundance, size distribution, and biological origins. Limnol. Oceanogr. 43, 536–542 (1998).

Article  ADS  CAS  Google Scholar 

Cullen, J. J. Subsurface chlorophyll maximum layers: Enduring enigma or mystery solved?. Annu. Rev. Mar. Sci. 7, 207–239 (2015).

Yoshida, A. et al. Abundances and microphysical properties of light-absorbing iron oxide and black carbon aerosols over east Asia and the arctic. J. Geophys. Res. 125, e2019JD032301 (2020).

Article  ADS  CAS  Google Scholar 

Moulin, A. J., Moum, J. N., Shroyer, E. L. & Hoecker-Martínez, M. Freshwater lens fronts propagating as buoyant gravity currents in the equatorial Indian Ocean. J. Geophys. Res. Oceans https://doi.org/10.1029/2021JC017186 (2021).

Kawai, Y. & Wada, A. Diurnal sea surface temperature variation and its impact on the atmosphere and ocean: A review. J. Oceanogr. 63, 721–744 (2007).

Weinbauer, M., Cattaneo, R., Malits, A. & Rassoulzadegan, F. Black carbon and microorganisms in aquatic systems. Adv. Environ. Res. 25, 1–37 (2012).

Wells, M. L. & Goldberg, E. D. Occurrence of small colloids in sea water. Nature 353, 342–344 (1991).

Article  ADS  CAS  Google Scholar 

Dunne, J. P., Sarmiento, J. L. & Gnanadesikan, A. A synthesis of global particle export from the surface ocean and cycling through the ocean interior and on the seafloor. Glob. Biogeochem. Cycles https://doi.org/10.1029/2006GB002907 (2007).

Tréguer, P. et al. Influence of diatom diversity on the ocean biological carbon pump. Nat. Geosci. 11, 27–37 (2018).

Schulz, M. et al. Atmospheric transport and deposition of mineral dust to the ocean: Implications for research needs. Environ. Sci. Technol. 46, 10390–10404 (2012).

Article  ADS  CAS  PubMed  Google Scholar 

Luo, C. et al. Combustion iron distribution and deposition. Glob. Biogeochem. Cycles https://doi.org/10.1029/2007GB002964 (2008).

This work was supported by a Grant-in-Aid for Scientific Research on Innovative Areas 6102 from JSPS of Japan, JSPS KAKENHI (JP19H05699, JP23KJ2144), the Environment Research and Technology Development Fund (JPMEERF20202003, JPMEERF20232001) of the Environmental Restoration and Conservation Agency provided by Ministry of the Environment of Japan, and the Arctic Challenge for Sustainability II (ArCS II) (JPMXD1420318865). We thank Takuma Miyakawa, the crew of R/V Shinsei-maru, and all persons involved in the cruise for their support. We also thank Kumiko Goto-Azuma, Hiromi Okumura, and Kaori Fukuda for analyzing SPM samples using Coulter Counter. We also extend our gratitude to Ryosuke Makabe and Keigo D. Takahashi for providing valuable insights about biological SPM in the ocean.

National Institute of Polar Research, Tachikawa, Tokyo, Japan

Atsushi Yoshida & Yutaka Tobo

Graduate Institute for Advanced Studies, SOKENDAI, Tachikawa, Tokyo, Japan

Meteorological Research Institute, Tsukuba, Ibaraki, Japan

Tokyo Metropolitan University, Hachioji, Tokyo, Japan

Japan Agency for Marine-Earth Science and Technology, Yokosuka, Kanagawa, Japan

Yoshimi Kawai & Kosei Sasaoka

The University of Tokyo, Tokyo, Japan

You can also search for this author in PubMed  Google Scholar

You can also search for this author in PubMed  Google Scholar

You can also search for this author in PubMed  Google Scholar

You can also search for this author in PubMed  Google Scholar

You can also search for this author in PubMed  Google Scholar

You can also search for this author in PubMed  Google Scholar

You can also search for this author in PubMed  Google Scholar

A.Y. and T.Y designed the research. A.Y. collected the seawater samples. A.Y. performed CAS analyses of all SPM samples. K.A. performed SEM analyses of CAS34 and CAS35 samples. K.S. and Y.K. performed chlorophyll analysis. A.Y. and Y.T analyzed the results. Y.K. and M.K. administrated the research cruise. N.M. developed the CAS instrument. All authors reviewed the manuscript.

The authors declare no competing interests.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Yoshida, A., Tobo, Y., Adachi, K. et al. Analysis of oceanic suspended particulate matter in the western North Pacific using the complex amplitude sensor. Sci Rep 14, 20055 (2024). https://doi.org/10.1038/s41598-024-70683-1

DOI: https://doi.org/10.1038/s41598-024-70683-1

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Scientific Reports (Sci Rep) ISSN 2045-2322 (online)

dammann gratings Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.