Skip to main content
NCBI home page
Scientific Reports logoLink to Scientific Reports
. 2024 Nov 4;14:26582. doi: 10.1038/s41598-024-74818-2

Statistical analysis of the association between El Niño and the biological carbon pump in the East Sea (Japan Sea)

Geunsoo Jang 1,#, Seunghyun Hong 2,#, Janghun Oh 2, Young-Il Kim 3, Minkyoung Kim 4,✉,#, Hyojung Lee 2,✉,#
PMCID: PMC11535021  PMID: 39496640

Abstract

Understanding the impacts of climate change on oceanic carbon cycling is important from a carbon sequestration perspective. A sediment trap study focused on the biological carbon pump system in the Ulleung Basin (UB) in the southwestern part of the East Sea (Japan Sea) was conducted from 2011 to 2017. Particulate organic carbon (POC) flux significantly increased by 37, 56, and 43% from 2014 to 2016 during the El Niño phase. We examined data related to water current variability, such as sea surface height, current velocity, and eddy frequency to understand their roles in particle transport. In addition, a Martin curve was employed to analyze the rate of vertical attenuation of POC flux in the UB. Current variability could be the most important factor influencing the increase in sinking-particle flux during the El Niño phase.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-024-74818-2.

Keywords: Particulate organic carbon, Biological carbon pump, Statistic analysis, Eddy, Lateral transport

Subject terms: Environmental sciences, Ocean sciences

Introduction

The biological carbon pump (BCP) is one of the main mechanisms of the oceanic carbon pump system. It involves a series of biogeochemical processes associated with carbon sequestration1. In brief, primary production (PP) in the upper layers of the ocean water column fixes inorganic carbon into biomass. These particles, including particulate organic carbon (POC), start to sink into the deep ocean interior, undergo remineralization, leading to attenuation with depth, and are stored in deep-sea sediments2. Accordingly, the sinking POC flux is primarily affected by the seasonal variation of PP. In addition to this, aeolian deposition, riverine inputs, sediment resuspension, and lateral transport can affect the sinking POC flux3–7. Mesoscale dynamics such as eddies, and episodic typhoons can also affect the POC flux8–10. Climate changes, such as global warming, potentially directly and indirectly impact the oceanic uptake of atmospheric CO2; therefore, understanding its dynamics and controlling factors is important11. Previous studies revealed that El Niño can be linked to global atmospheric changes triggered by fluctuations in sea surface temperatures (SST)12.

Additionally, variations in sea surface wind patterns and ocean water depth may influence the occurrence of El Niño13. Furthermore, significant climate changes can persist even after the disappearance of El Niño14. El Niño has different characteristics depending on where it occurs15, and it has different impacts depending on the ocean14,16–20 and the stage of El Niño21,22.

Sediment traps have been widely used to collect sinking particles in the ocean interior to better understand BCP23,24. However, long-term sediment trap studies are globally restricted due to their operation and maintenance difficulties. Therefore, statistical methods are crucial in Earth and Ocean Sciences for addressing the challenges posed by limitations in field observations, difficulties in data collection, and the often-encountered issue of data scarcity25–27.

The East Sea (Japan Sea, hereafter EJS) is a semi-enclosed marginal sea located east of the Korean Peninsula, surrounded by Russia and the Japanese Islands. Physical and biogeochemical changes in the EJS are known to occur faster than those in the adjacent Pacific; therefore, long-term and multidisciplinary approaches to understanding changing biogeochemical parameters are necessary28. Continuous sediment trap deployment has been conducted in the Ulleung Basin (UB) in the southwestern part of the EJS since 201129,30. Kim et al.28 examined the 1-year sinking-particle flux at 1000 m and 2000 m depths in the UB from 2011 to 2012, showing a bimodal POC flux with the highest flux in spring, reflecting variations in net PP. The following study by Kim et al.29 expanded the idea and deployed sediment traps at 500 m, 1000 m, and 2000 m depths to collect sinking-particle samples from 2011 to 2017.

Interestingly, the sinking POC fluxes in the UB significantly increased during the El Niño phase (2014–2016)29. The increase in POC flux was interpreted in conjunction with factors such as PP, aeolian deposition, and the Oceanic Niño Index (ONI). In particular, mesoscale eddies and their role in transporting resuspended sediments have been proposed. However, the important factors have not been clearly described, and we recognize the need to investigate these aspects further by assessing the statistical analyses in our study. Further research and data analyses are required to better understand the relative contributions of these factors to the increased particle flux in the UB.

In the present study, we aimed to analyze the impact of two factors on the increase in POC flux in the UB 2014 to 2016: (i) the association with the 2015/16 El Niño and (ii) enhanced particle transport resulting from current variability on increased sinking POC fluxes in the UB. Elevated SST during El Niño can increase surface productivity, which results in increased sinking POC. In addition, the increase in sinking POC flux also represents an increase in export efficiency (e.g., the POC/PP ratio). Surface current variability can partly explain the influences of mesoscale physical dynamics on sinking particles. We categorized the periods into El Niño and non-El Niño phases and performed a thorough statistical analysis to explore the factors behind the increase in particle flux. Our comprehensive statistical analysis identified the most influential factor contributing to the increase in particle flux during the El Niño phase. We anticipate that this interdisciplinary analysis has the potential to provide valuable insights into the interpretation of results in the field of Earth and Ocean Sciences.

Results

Association between the impact of the SST and El Niño phase

We examined the impact of SST increases on POC flux by dividing the dataset into El Niño and non-El Niño phases. Figure 1 shows the POC flux at depths of 500 m, 1000 m, and 2000 m from July 2011 to September 2017. The POC flux during the El Niño phase was higher than that during the non-El Niño phase (Fig. 130), . The difference was particularly pronounced between 500 m and 1000 m depths, and even at 2000 m, differences in POC flux were evident, excluding non-observed data. We observed an elevation in POC flux at 500 m, 1000 m, and 2000 m during El Niño. Typically, POC flux tends to rise with an increase in PP or chlorophyll-a (Chl-a). However, Fig. 1E indicates that Chl-a did not exhibit a significant increase during El Niño, except for a sudden spike in Chl-a in 2015 at a specific point; this implies that the rise in POC flux is not attributed to increased productivity during El Niño.

Fig. 1.

Fig. 1

Open in a new tab

Sinking POC flux, ONI, Chlorophyll-a, and primary production data from July 2011 to September 2017. (A-C) The POC flux at 500 m, 1000 m, and 2000 m, respectively; (D) ONI, (E) chlorophyll-a (Chl-a), and (F) primary production (PP) from July 2011 to September 2017. The red area indicates the El Niño phase from April 2014 to May 2016.

Table 1 provides a summary of the particle flux and current variability data across different phases, comparing all phases, non-El Niño phases, and El Niño phases. It includes key parameters such as Chl-a, PP, and POC fluxes at various depths (500 m, 1000 m, 2000 m), SST, current velocity (VEL), and sea surface height (SSH). This comparison helps to understand the variations in oceanographic conditions during different climate phases.

Table 1.

Summary of particle flux and current variability data.

Parameters All phases Non-El Niño phase El Niño phase U statistic
(P-value)
Chl-a (mg/m3) 0.59 Inline graphic0.39

0.58 Inline graphic0.37

(n = 70)

0.66Inline graphic0.46

(n = 26)

770,138***

(< 0.01)
PP (mgC/m2d) 793 Inline graphic292.67

775Inline graphic257.89

(n = 70)

899 Inline graphic344.18

(n = 26)

749,086***

(< 0.01)
500 m (mgC/m2d) 28 Inline graphic23.80

21 Inline graphic18.41

(n = 82)

37 Inline graphic26.86

(n = 61)

36,505***

(< 0.01)
1000 m (mgC/m2d) 26 Inline graphic16.46

18 Inline graphic7.91

(n = 65)

36 Inline graphic18.87

(n = 54)

47,234***

(< 0.01)
2000 m (mgC/m2d) 28 Inline graphic16.32

22 Inline graphic10.76

(n = 95)

40 Inline graphic19.54

(n = 45)

49,857***

(< 0.01)
SST (°C) 17.1 Inline graphic5.39

17.4 Inline graphic5.62

(n = 2222)

16.4 Inline graphic4.63

(n = 792)

539,928***

(< 0.01)
VEL (m/s) 0.21 Inline graphic0.10

0.21 Inline graphic0.11

(n = 2222)

0.25 Inline graphic0.09

(n = 792)

744,342

(1.4)
SSH (m) 0.16 Inline graphic0.10

0.16 Inline graphic0.11

(n = 2222)

0.16 Inline graphic0.07

(n = 792)

744,579

(1.19)

Frequency of ACE

(event/month*)

 73.2 Inline graphic27.03 77.37 Inline graphic27.57 65.34 Inline graphic24.57

487

(0.1)
Amplitude of ACE (m) 0.08 Inline graphic0.05 0.07 Inline graphic0.05 0.10 Inline graphic0.06

965***

(< 0.01)
Frequency (event/month) of CE 73.16 Inline graphic26.25 71.47 Inline graphic26.41 76.35 Inline graphic26.17

698.5

(0.5)
Amplitude of CE (m) 0.04Inline graphic0.03 0.04Inline graphic0.03 0.04Inline graphic0.03

532

(0.24)
Open in a new tab

*Event/month refers to the number of occurrences of eddy per month.

Figure 2 compares the monthly POC fluxes between the El Niño and non-El Niño phases at 500 m, 1000 m, and 2000 m. While SST showed little difference between the El Niño and non-El Niño phases, it can be noted that both Chl-a and PP exhibited a sudden peak during the El Niño phase, particularly in April and May, accompanied by substantial variability. Therefore, it can be concluded that the El Niño event that occurred during 2015/2016 is not significantly related to the SST or carbon production in the surface water. The increase in POC flux during the El Niño phase cannot be solely attributed to changes at the sea surface. Other factors, such as lateral transport of POC from different regions or resuspension likely played a significant role in the observed increase in POC flux. For example, in Fig. 2C and E, and 2F, we observed that during a specific period (April-August), PP and Chl-a were higher during the El Niño phase than the non-El Niño phase. However, the POC flux at 2000 m was consistently higher during the El Niño phase.

Fig. 2.

Fig. 2

Open in a new tab

Monthly comparison of factors related to the particle flux between El Niño and non-El Niño phase. (A) sea surface temperature (SST), (C) primary production (PP), (E) chlorophyll-a (Chl-a), The POC flux at 500 m, 1000 m, and 2000 m depth for (B), (D), and (F). The red line indicates the El Niño phase, and the blue line indicates the non-El Niño phase. The error bars indicate the mean ± standard deviation. No line means that there is no available data.

Furthermore, the Mann-Whitney U test was employed to determine if there are variations in the monthly rank sum values of different oceanographic variables during El Niño and non-El Niño phases. In Table 2, we observed significant statistical differences in PP from April to October and in December, as well as significant differences in Chl-a were noted from May to October and in December. In contrast, at a depth of 1000 m, the POC flux showed significant statistical differences almost consistently, with greater quantities noted during El Niño periods; this indicates that other factors beyond the slightly lower SST during El Niño might influence the observed differences.

Table 2.

Summary of Mann-Whitney U test of POC flux at 500 m, 1000 m, and 2000 m depths, PP, and Chl-a.

Month POC flux at 500 m POC flux at 1000 m POC flux at 2000 m PP Chl-a
1

4425

(0.073)

 4746.5*** (< 0.01)

1054***

(< 0.01)

4031

(0.589)

3410

(0.209)
2

4213***

(< 0.01)

 4381*** (< 0.01)

2052***

(< 0.01)

2814

(0.179)

2814

(0.179)
3

5002***

(< 0.01)

4506*

(0.036)

4200.5

(0.181)

3844

(1)

3699

(0.675)
4

5370***

(< 0.01)

4456

(0.232)

5711***

(< 0.01)

6268***

(< 0.01)

4576

(0.131)
5

6207***

(< 0.01)

4133.5

(0.586)

6727***

(< 0.01)

7347***

(< 0.01)

6045***

(< 0.01)
6

5925***

(< 0.01)

 4883*** (< 0.01)

6714***

(< 0.01)

6304***

(< 0.01)

5884***

(< 0.01)
7

5625.5*

(0.012)

 6205.5*** (< 0.01)

8480***

(< 0.01)

8556***

(< 0.01)

9610***

(< 0.01)
8

3530.5***

(< 0.01)

4660

(0.609)

7789.5***

(< 0.01)

8215***

(< 0.01)

8556***

(< 0.01)
9

4127

(0.503)

4850

(0.077)

3117.5***

(< 0.01)

6660***

(< 0.01)

5760***

(< 0.01)
10

5177***

(< 0.01)

 5285.5*** (< 0.01)

1922***

(< 0.01)

2552***

(< 0.01)

2594***

(< 0.01)
11

5550***

(< 0.01)

 5438*** (< 0.01)

2088***

(< 0.01)

4232

(0.054)

4008

(0.215)
12

3844

(1)

3900

(0.864)

961***

(< 0.01)

 4866*** (< 0.01)

4577*

(0.033)
Total

730,722***

(< 0.01)

 730,722*** (< 0.01)

615,990*

(0.036)

749,086***

(< 0.01)

 770,138*** (< 0.01)
Open in a new tab

Note: * indicates statistical significance at alpha 0.05 level; *** indicates statistical significance at alpha 0.01.

Association between the impact of the surface current variability and the El Niño phase

We investigated the impact of current variability in surface water during the El Niño phase on sinking POC flux. Figure 3 compares the monthly SSH, VEL, ACE, and CE averages between the El Niño and non-El Niño phases. Figure 3B and D show that eddy current velocity (m/s) and amplitude (m) were somewhat higher during the El Niño phase, from January to March and June to August. When comparing the El Niño and non-El Niño phases, the frequency (event/month) of eddies did not exhibit statistically significant differences. Although no significant differences were observed in the frequency of CE and ACE between El Niño and non-El Niño periods, our investigation revealed that the average amplitude of ACE was consistently higher during El Niño events. This finding suggests that the intensity of these eddies, rather than their mere occurrence, may play a critical role in modulating the observed patterns of POC flux. Even though ACE frequency between El Niño and non-El Niño phases were not different, it can be interpreted that the monthly average amplitude of eddies was notably higher.

Fig. 3.

Fig. 3

Open in a new tab

Monthly comparison of factors related to the current variability between El Niño and non-El Niño phases. (A) Sea surface height (SSH) (B) Velocity of sea surface current (VEL) (C) and (E) The frequency of anticyclonic eddy (ACE) and cyclonic eddy (CE), where event/month refers to the number of occurrences of eddy per month. (D) and (F) The amplitude (m) of ACE and CE. The red line indicates the El Niño phase, and the blue line indicates the non-El Niño phase. The error bars indicate the mean ± standard deviation. No line means that there is no available data.

In addition, Fig. 4 compares the new occurrence, speed, and amplitude of ACE and CE between El Niño and non-El Niño phases. Figure 4A and C illustrate the occurrence of ACE and CE, whereas Fig. 4B and D represent the amplitude and speed of ACE and CE. The circle size denotes the radius of eddy, and the color gradient from white to red or blue indicates speed values, with darker colors representing faster eddy speeds. The amplitude of CE is similar between El Niño and non-El Niño periods, whereas the amplitude of ACE is greater during El Niño periods. Additionally, ACE has a larger radius and faster speeds. It could be concluded that eddies with higher amplitude play a significant role in the observed increase in sinking POC flux during the El Niño phase, indicating that the intensity of these eddies is a crucial factor influencing the sinking POC flux patterns.

Fig. 4.

Fig. 4

Open in a new tab

Monthly comparison of eddy factors between El Niño and non-El Niño phases. (A) and (C) show the monthly occurrence of cyclonic eddy (CE) and anticyclonic eddy (ACE). (B) and (D) show the amplitude, speed, and radius of CE and ACE. The color gradient from white to red or blue indicates speed values. The circle size denotes the radius of eddy.

We applied the SARIMAX model to analyze whether ENSO influences the time-series patterns of various oceanographic physical variables. While the Mann-Whitney U test provided an initial assessment of the differences in monthly distributions of these variables between El Niño and non-El Niño phases, it was not thorough enough to definitively determine variations in their time-series patterns. The SARIMAX model included a dummy variable for El Niño phases to assess its significance, aiming to statistically ascertain if the time-series patterns of oceanographic variables change during the El Niño phase, as shown in Table S2. As a result, we found significant differences in the time-series patterns of the ACE amplitude and speed average variables between El Niño and non-El Niño periods.

POC export efficiency by analyzing the Martin curve

POC export efficiency by different depth

For a better understanding of differences in POC export efficiencies during the non-El Niño and El Niño phases, we adopted a Martin curve equation. We extracted five time periods for the estimation of the curve from the sinking POC flux data that were continuously obtained at 500 m, 1000 m, and 2000 m; from July 4, 2011, to November 7, 2011 (Inline graphic, December 1, 2012, to June 27, 2013 (Inline graphic); June 1, 2014, to August 9, 2014 (Inline graphic); March 15, 2015, to June 6, 2015 (Inline graphic); and April 10, 2015, to May 10, 2015 (Inline graphic), as shown in Fig. 5. Adhering to ONI > 0, Inline graphic and Inline graphic corresponded to the non-El Niño phase. Inline graphic, Inline graphic, and Inline graphic are El Niño phases. We observed that the sinking POC flux at 1000 m and 2000 m were much larger during the El Niño phase, as shown in Table S3. The mean sinking POC flux at 1000 m in Inline graphic were 25 ± 7.90 (mgC/m2d) and the mean of sinking POC flux at 1000 m in Inline graphic were 45 ± 22.47 (mgC/m2d). Similarly, Table S3 shows higher values for the El Niño phase than for the non-El Niño phase.

Fig. 5.

Fig. 5

Open in a new tab

Periods fromInline graphictoInline graphicfor the Martin curve. (A-C) POC flux at 500 m, 1000 m, and 2000 m, respectively; (D) POC flux at 2000 m for five time periods. The blue shaded area indicates the non-El Niño phase, and the red shaded area indicates the El Niño phase.

POC export efficiency by using the Martin curve

Figure 6A–E compare the estimated POC export efficiency using the mean of POC flux data across depths (500 –2000 m) during Inline graphic periods. During the non-El Niño phase (Inline graphic), the POC flux decreased with increasing depth from 500 m to 2000 m. However, during the El Niño phase (Inline graphic), there were instances where the POC flux slightly decreased or increased with depth.

Fig. 6.

Fig. 6

Open in a new tab

Comparison between the POC flux data and estimated Martin curve. Circle points indicate the mean of the POC flux data. Error bars represent the mean ± standard deviation of the observed data. Inline graphicrepresents the average POC flux data at 500 m. Inline graphic indicates the rate of POC flux attenuation. Dashed lines indicate the estimated Martin curve using Inline graphic and Inline graphic for each period from Inline graphic to Inline graphic: (A) Inline graphic period (B) Inline graphic period (C) Inline graphic period (D) Inline graphic period (E) Inline graphic period. (F) POC flux data for the non-El Niño phase. Solid lines indicate the estimated Martin curve for the non-El Niño phase (Inline graphic): (G) non-El Niño phase. Green, purple, and yellow areas represent a positive anomaly regarding the mass of POC transported per meter per day: (H) Inline graphic period (I) Inline graphic period (J) Inline graphic period.

This variability can be explained by factors such as sediment resuspension and/or lateral transport in the UB, as discussed by28,29,31. It is important to note that not all POC at 2000 m depth is newly produced28,29. Despite the anticipated higher POC flux at 2000 m due to water depth (~ 2300 m) and dynamic environmental conditions, it is important to note that not all POC at this depth is newly produced.

We defined Inline graphic as the average POC flux data at 500 m and mathematically obtained the better fit with b values for each measured POC flux data for each period of Inline graphic (Fig. 6A-E), as shown in Table 3. The parameter Inline graphic indicates the rate of POC flux attenuation from Inline graphic and is interpreted as the BCP efficiency. The parameter Inline graphic was 0.4869, 0.2774, 0.0495, 0.1037, and − 0.3740 for each fitting curve. The estimated Inline graphic was higher during the non-El Niño phase than during the El Niño phase, which indicates that vertical attenuation of sinking POC was higher during the non-El Niño phase than during the El Niño phase. POC flux for Inline graphic and Inline graphic at 500 m (Inline graphic) was found to be similar, with values of 41.26 (Inline graphic and 39.72 (Inline graphic, respectively. However, the POC flux vertical attenuation rates differed significantly. For Inline graphic, the POC flux at 2000 m (Inline graphic) was estimated at 21.01, representing a substantial 49% decrease compared to Inline graphic. However, Inline graphic, the POC flux at 2000 m (Inline graphic) was estimated at 37.08 (Inline graphic, indicating a more modest 7% decrease compared with Inline graphic. Moreover, POC fluxes during Inline graphic at 2000 m (Inline graphic) increased remarkably compared to that with Inline graphic. It could be interpreted that besides the decrease in vertical attenuation, the influence of the lateral transport of POC could also be a contributing factor.

Table 3.

Estimation of parameters for the Martin curve and the ratio of POC flux to PP for each fitting period.Inline graphic is the attenuation parameter. Inline graphic, Inline graphic, Inline graphic are POC flux at 500 m, 1000 m, 2000 m. Inline graphic and Inline graphic are estimated.

non-El Niño El Niño
Inline graphic
(Fig. 6A)		 Inline graphic
(Fig. 6B)		 Inline graphic
(Fig. 6G)		 Inline graphic
(Fig. 6C)		 Inline graphic
(Fig. 6D)		 Inline graphic
(Fig. 6F)
Parameters of Martin curve
 Inline graphic 0.4869 0.2774 0.3669 0.0495 0.1037 -0.3740
 Inline graphic 41.26 (-) 20.91 (-) 26.71 (-) 39.72 (-) 30.14 (-) 29.59 (-)
 Inline graphic 29.44 (-29%) 17.25 (-17%) 20.72 (-22%) 38.38 (-3%) 28.05 (-7%) 38.35 (+ 30%)
 Inline graphic 21.01 (-49%) 14.23 (-32%) 16.06 (-60%) 37.08 (-7%) 26.10 (-13%) 49.70 (+ 68%)
e-ratio Inline graphic
 500 m 0.053 (-) 0.029 (-) 0.037 0.037 (-) 0.021 (-) 0.030 (-)
 1000 m 0.032 (-39%) 0.023 (-21%) 0.026 (-30%) 0.040 (+ 7%) 0.021 (+ 4%) 0.046 (+ 52%)
 2000 m 0.029 (-45%) 0.020 (-31%) 0.023 (-28%) 0.035 (-7%) 0.018 (-14%) 0.053 (+ 77%)
Open in a new tab

We estimated the attenuation rate during the non-El Niño phase to identify the anomaly regarding the mass of POC transported during the El Niño. Figure 6F and G show POC flux and the estimated attenuation rate for the non-El Niño phase, respectively. As a result, the rate of POC flux attenuation Inline graphic was estimated at 0.3669. Figure 6H-J compared the mean of the POC flux data concerning the estimated POC flux using the b values 0.3669 at Inline graphic, Inline graphic, and Inline graphic, respectively. Green, purple, and yellow areas represent a positive anomaly regarding the mass of POC transported per meter square per day in Fig. 6H-J.

In Table 3, the ratio of the mean POC flux to PP (i.e., e-ratio) by depth is compared for different periods. During the non-El Niño phase (Inline graphic), the ratio decreased as depth increased, which is a general trend32. Nonetheless, during the El Niño phase, the increasement at 1000 m as 0.040 (+ 7%), 0.021 (+ 4%), and 0.046 (+ 52%) for Inline graphic, Inline graphic, and Inline graphic, respectively. These findings could be interpreted as an external input of POC occurring within a depth range of 500 m to 1000 m during El Niño.

Discussion

Kim et al.29 observed increased sinking POC flux during the El Niño phase in the UB and proposed the potential effect of mesoscale eddies. Other studies have shown that mesoscale eddies can affect the seasonal variability of BCP systems17,33,34. Gaube et al.33 found that nonlinear mesoscale eddies can enhance Chl-a and phytoplankton biomass by up to 100% through Ekman pumping, which impacts the PP and marine ecosystems. Dufois et al.34 found that eddies in the South Indian Ocean affect the surface Chl-a during winter by seasonal adjustment of the mixed layer depth by eddies. The effects of El Niño on sinking-particle flux in the UB are poorly understood, whereas those in the Pacific have been relatively well studied35–37. A continuous, long-term observational study is necessary to better understand the association of the BCP with climate change in the UB. With oceanic sample collection and observations, it is beneficial to adapt mathematical and statistical analyses to compensate for data deficiencies.

This study investigated the factors responsible for the notable increase in sinking POC flux during the El Niño phase from 2014 to 2016. The increase in POC flux indicates (i) an increase in PP and/or (ii) an increase in carbon export efficiency, and/or (iii) carbon transport from elsewhere. Substantial increases in Chl-a and PP were observed during the El Niño phase, particularly in April and August. To address this, we compared SST changes between the El Niño and non-El Niño phases to clarify the influence of the increase in POC flux. As a result, it became apparent that the upsurge in sinking POC flux during El Niño could not be solely attributed to SST changes when comparing the El Niño and non-El Niño phases. This finding suggests that a more intricate interplay of factors is involved, warranting further research.

Additionally, we investigated the impact of current variability during El Niño phases on sinking POC flux. We compared the monthly SSH, VEL, and eddies (ACE and CE) averages between El Niño and non-El Niño phases. Although most variables did not exhibit distinct time-series patterns during El Niño, the current velocity showed a notable increase, particularly from January to March and June to August, with significant variations from February to March. Current variability can influence the sinking POC flux dynamics, particularly in current velocity and eddy amplitude. Using the Mann-Whitney U test, we found that the differences observed might be due to factors other than increased SST during El Niño.

Additionally, by applying the SARIMAX model to assess the impact of ENSO on time-series patterns of various oceanographic physical variables, we identified significant differences in the time-series patterns, specifically in the amplitude and speed average variables of ACE, between El Niño and non-El Niño phases; this suggests that the amplitude of ACE was greater during El Niño phase than in non-El Niño phase. This increase in ACE could be a key factor contributing to the lateral transport of particles, as inferred from the analysis of eddy data.

We utilized the Martin curve to predict the estimated anomaly regarding the mass of POC transported and compare the POC attenuation rates during the El Niño and non-El Niño phases to further evaluate the extent of current variability. We found that the laterally transported particles at 1000 m and 2000 m were notably higher during the El Niño phases. Analyzing five distinct periods, the attenuation rate (b) was greater during non-El Niño phases, with a particularly intriguing negative Inline graphicvalue at -0.3740, indicating that the cessation of POC flux increased with depth. Surface ocean dynamics possibly influence the POC flux at 500 m. However, Due to current variations, the augmented particle flux may be attributed to intensified particle movement and transport to the deep ocean interior. At 1000 m and 2000 m depths, the change of POC flux is influenced by current variability. A comparison of the POC fluxes at 500 m (Inline graphic) and 2000 m (Inline graphic) revealed significant differences in the attenuation rates between the two phases, underscoring the influence of lateral influx during El Niño. As a result, we statistically analyzed the data to demonstrate that lateral particle transport is a more likely significant factor influencing the increase in POC flux compared to the rise in SST during the El Niño phase, aligning closely with previous suggestions in existing research30. This comparative analysis allowed us to thoroughly explore the factors contributing to the increase in the sinking POC flux and the significance of lateral transport.

Limitations of the study

Despite new efforts to combine oceanography and statistics, several limitations remain. Herein, we present the limitations and potential future studies. First, although our study thoroughly analyzed various factors related to the increase in sinking POC flux, including SST, SSH, VEL, and eddies, it is important to acknowledge that additional factors may be considered.

Second, relevant data are lacking. This study reused published sinking particle data, and the sampling period was sufficiently short to cover long-term variations. Further deployment of mooring lines with sediment traps and current meters is necessary to gain insights into the relationship between ENSO and its effect on the BCP. Biologically, mesoscale eddies can supply nutrients and modulate light availability in the surface water, thereby enhancing PP38. To thoroughly understand this process, combined research cruises to collect water samples and measure nutrients in depth are necessary. The current study has limitations as it did not encompass all potential contributing elements. The study site in the East Sea (Japan Sea) exhibits complex physical dynamics, making it challenging to trace particle sources using the surface current data alone due to the absence of a current meter deployment in this mooring line. To address this, various biogeochemical tracers have been utilized to understand the sources and trajectories of sinking particles30,31,39. Regardless, it is essential to recognize that this research augmented its analysis by supplementing oceanographic data, such as velocity and eddy data, and employing mathematical and statistical approaches to quantitatively compare the POC attenuation levels. These additional statistical analyses add significance to our investigation of the abrupt increases in sinking POC flux.

Third, we did not compare the POC flux during several El Niño phases in various regions. Studies about El Niño show that it has different characteristics depending on where it occurs15 and has different impacts depending on the ocean14,16–20 and the stage of El Niño21. In addition, few studies link El Niño to sinking particles. The previous research in the subtropical Pacific region36,37 highlighted the significant impact of ENSO events. However, the sampling periods for these studies span from 2003 to 2013, which do not coincide with our sampling periods.

Furthermore, we conducted a Martin curve fitting process with data from specific depths (500 m, 1000 m, and 2000 m). Data from a broader range of depths could have provided more extensive insights, which could be further solved by collecting high-resolution in-depth water with large-volume in situ pumps at the same site. Furthermore, to ensure data consistency across all three depths, we limited our fitting to five distinct periods: Inline graphic to Inline graphic and accordingly, the seasonal variations in sinking particle fluxes and compositions that might affect the sinking particle decomposition could not be fully considered. Despite these limitations, we obtained sufficient results to examine the differences in sinking POC flux attenuation between El Niño and non-El Niño phases.

In conclusion, our study demonstrates the complex nature of the sinking POC dynamics in the UB, particularly during El Niño events. Although SST alone does not explain the observed increase in POC flux, factors such as current variability in the surface water and the influence of eddies appear to play a significant role. Martin curve analysis further substantiated the influence of El Niño on the attenuation of sinking POC flux. The importance of our study lies in the comprehensive examination of these factors and their combined impact on the sinking POC flux. This reanalysis proposes the necessity of multidisciplinary approaches in ocean and earth sciences with a lack of available data. These findings contribute to a better understanding of the intricate processes governing carbon sequestration in the ocean, with implications for climate change and ocean carbon cycling studies. It can also help us comprehend how climate change impacts oceanic carbon flux, providing valuable data for future circulation change predictions.

Methods

Regional setting

The sinking-particle data used in this study were obtained from Kim et al. (2020). In brief, sinking particles intercepted by bottom-tethered sediment traps deployed on a mooring at three depths (500 m, 1000 m, and 2000 m; ∼2370 m total water depth) between 2011 and 2017 at station ES (37.63°N, 131.36°E) in UB (Fig. 7). Further details on the sampling method are described in Kim et al.29. The PP and Chl-a in this study obtained from the previous study30. The PP was estimated using the Vertically Generalized Productivity Model (VGPM)38 and the optimal carbon fixation rate31,40. For the statistical analysis, we utilized PP and the Chl-a concentration. Although the existence of the subsurface chlorophyll maximum (SCM) may increase the uncertainty in the satellite-derived NPP, in the EJS, the Moderate-Resolution Imaging Spectroradiometer (MODIS)-derived PP41,42 is relatively consistent with in situ measurements of PP43. Therefore, we decided to use the satellite data to enlarge the number of data points for a more robust statistical analysis. The ONI represents the 3-month moving average of the monthly SST anomalies in the Niño 3.4 region, a monitoring area for El Niño/La Niña events44. It serves as an indicator of the El Niño phase or the non-El Niño phase. We extracted oceanic physical data, including SST, SSH, and VEL, from Mercator Ocean’s45 analysis from July 1, 2011, to September 30, 2017. These data were collected within the geographical bounds of latitude 33.75°–38.25°N and longitude 125.25°–134°E (Fig. 7). We calculated the averages of these variables in the vicinity of the four specified areas marked A, B, C, and D in Fig. 7, originating from a specific location marked with an asterisk. This specific location served as the mooring site for obtaining sample data. Thus, we selected three oceanic physical variables–SSH, SST, and VEL–obtained from the Copernicus Marine Service45. We obtained the frequency and magnitude of mesoscale eddies using eddy tracking data in Aviso + to match the spatial and temporal domains of the oceanic physical data46. Eddies are detected using Absolute Dynamic Topography maps generated using the most advanced altimeter satellites. This study found that the primary vertical mechanism influenced the differences between anticyclonic eddies (ACE) and cyclonic eddies (CE) during peak phytoplankton biomass in winter.

Fig. 7.

Fig. 7

Open in a new tab

Map of the study site, UB. The left panel shows the average sea surface height (SSH; color) and velocity of the sea surface current (VEL; arrows) during the El Niño phase for the entire spatial domain of the oceanic physical variables and mesoscale eddy dataset, which is located between 33.75°N and 38.25°N latitude and 125.25°E and 134°E longitude. The blown-up figure on the right indicates the four grid cells (A, B, C, and D) with a spatial resolution of 0.25 degrees near the sediment trap mooring site (red asterisk).

Materials and methods

Analysis of two factors: increased POC flux during El Niño

The ONI was used to classify the El Niño events, which indicates the influence of El Niño-Southern Oscillation (ENSO) on the Pacific climate. According to the Climate Prediction Center (CPC), El Niño is defined when the ONI exceeds 0.5 for five consecutive months44. To clarify the difference in POC flux between El Niño and non-El Niño phases, we defined the El Niño phase when the ONI was greater than zero (i.e., ONI > 0), which was also used in Lee et al.47. In addition, the Pacific Decadal Oscillation (PDO) also plays a significant role, especially in the North Pacific (north of 20°N). ONI is a better indicator to indicate EI Niño events than PDO even though the UB is not located in the equatorial Pacific region, where the influence effect of the EI Niño expected. We especially use the ONI as the high sinking particle flux periods match the super EI Niño phase well. There are some studies about the relationships between East Korea Warm Current and the EI Niño48–50.

According to the redefined definition of ONI and PDO, the period from April 2014 to May 2016 was the El Niño phase, as shown in Figure S1. We compared whether there were statistically significant differences between the El Niño and non-El Niño phases in terms of weekly or monthly averages of the variables.

Two factors have been proposed to explain the increased particle flux: one factor is the impact of the SST on PP, which is associated with the 2015/16 El Niño. The other factor is enhanced particle transport because of the current variability. Figure 8 shows the outline of the statistical analysis to explore the impact of two factors. Regarding the first factor, we examined surface productivity data, including Chl-a and PP, to investigate the productivity surge attributed to elevated SST during El Niño. Our analysis included data related to surface productivity, such as SST, Chl-a, and PP32,51. We compared the amounts of PP and the sinking POC during the El Niño and non-El Niño phases. For the second factor, we analyzed data related to current variability, including variables such as SSH, current velocity (VEL), and eddy occurrence frequency (Eddy), aiming to elucidate the potential causes of the heightened sinking POC flux. Furthermore, we employed the Martin curve to assess the differences in sinking POC flux attenuation during the El Niño and non-El Niño phases. These two factors collectively explain the increased particle flux from 2014 to 2016.

Fig. 8.

Fig. 8

Open in a new tab

Graphical abstract of two factors leading to an increase in POC flux. Factor 1 indicates the productivity surge attributed to elevated sea surface temperatures (SST). Factor 2 indicates the lateral transport by the current variability.

Vertical POC attenuation using Martin curve

It has been proposed that sinking POC in the UB is transported via dynamic sources, especially by using biogeochemical tracers such as radiocarbon values of POC29,30. With multidisciplinary specialties in the statistical analysis of various oceanic datasets, we analyzed several datasets and found that the sinking POC flux was more pronounced based on two primary factors.

We used the Martin curve as a mathematical model to re-examine these resuspended and/or laterally transported particles and quantitatively analyze the variability of the vertical POC flux in the UB. The Martin curve uses a power law to describe the POC flux attenuation2,3. Several studies have employed the Martin curve to investigate biogeochemical fluxes, particularly the PP and POC52,53. Cael et al.52 introduced a probabilistic approach, revealing that the probability distributions of two significant biogeochemical fluxes, the POC flux, and marine PP, follow a log-normal distribution. Furthermore, they demonstrated a statistically significant linear relationship between the two physical quantities53.

Efforts have been made to mathematically describe the relationship between the POC flux according to the water depth. Various models have been proposed, including those using exponential functions54, extended models such as the ballast model55, and models employing rational functions56. We employed the best-known and traditional model, the Martin curve2,57. In this study, we extend the Martin curve to 2000 m as follows:

graphic file with name M120.gif 1

where Inline graphic is the POC flux at a depth of Inline graphic (m) for Inline graphic, where z is the depth and Inline graphic is the baseline of the depth. We considered Inline graphic to be 500 m because the POC flux was available from the 500 m. In Eq. (1), Inline graphic indicates the rate of POC flux attenuation from baseline depth (Inline graphic). The parameter Inline graphic is interpreted as the BCP efficiency. A larger Inline graphic value corresponds to greater attenuation of the POC flux and lower efficiency of the BCP. We constructed Martin curves to fit the POC flux data denoted by Inline graphic for each location depth z (m). Inline graphic represent the POC flux data point Inline graphic at a depth of Inline graphic (m), where Inline graphic and Inline graphic is the total data at a Inline graphic (m). Due to the lack of data, we selected specific POC flux data periods observed for all three depths at 500 m, 1000 m, and 2000 m. We used a nonlinear least-squares method to estimate the parameters Inline graphic defined by:

graphic file with name M138.gif 2

In the present study, we described the data availability in the Regional setting session. All the data we analyzed are summarized in Table 1, S1, and S3.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1 (211.3KB, docx)

Acknowledgements

This research was supported by a project grant for the Government-wide R&D to Advance Infectious Disease Prevention and Control, Republic of Korea (No. HG23C1629). H. Lee was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (No. NRF-2022R1A5A1033624, NRF-2022R1C1C1006237). M. Kim is partly supported by the Research Program for the Carbon Cycle between Oceans, Land, and Atmosphere of the National Research Foundation (NRF) funded by the Ministry of Science and ICT (2022M3I6A1085990), and the Global-Learning & Academic Research Institution for Master’s, PhD students, and Postdocs (LAMP) Program of the Korea NRF grant funded by the Ministry of Education (No. RS-2023-00301914). G. Jang was supported by a National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (No. NRF-RS-2023-00274002).

Author contributions

G. Jang and S. Hong: Visualization, Validation, Formal analysis, Data curation, Writing – original draft, Methodology, J. Oh: Visualization, Formal analysis, Data curation, Y.I.Kim: Data curation, M. Kim : Writing – original draft, Conceptualization, H. Lee: Writing – review & editing, Writing – original draft, Supervision, Methodology, Conceptualization.

Data availability

All ocean sinking particle data used for this manuscript are available in the article/Supplementary Material of Kim et al. (2020; https://www.frontiersin.org/articles/10.3389/feart.2020.00144/full#supplementary-material). In the present study, we described the data availability in the Regional setting session. All the data we analyzed are summarized in Table 1, S1, and S2.

Declarations

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

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

These authors contributed equally: Geunsoo Jang, Seunghyeon Hong, Minkyoung Kim and Hyojung Lee.

Contributor Information

Minkyoung Kim, Email: minkyoung@knu.ac.kr.

Hyojung Lee, Email: hjlee@knu.ac.kr.

References

  • 1.Volk, T. & Hoffert, M.I. Ocean Carbon Pumps: Analysis of Relative Strengths and Efficiencies in Ocean-Driven Atmospheric CO2Changes. The Carbon Cycle and Atmospheric CO2: Natural Variations Archean to Present, vol. 32, 99–110. 10.1029/GM032p0099 (1985).
  • 2.Martin, J. H., Knauer, G. A., Karl, D. M. & Broenkow, W. W. VERTEX: Carbon cycling in the northeast Pacific. Deep Sea Res. A34, 267–285. 10.1016/0198-0149(87)90086-0 (1987). [Google Scholar]
  • 3.Shih, Y., et al. Elevated carbon flux in deep waters of the South China Sea. Sci. Rep.9 (2019). 10.1038/s41598-018-37726-w. [DOI] [PMC free article] [PubMed]
  • 4.Chung, Y. C. & Hung, G. W. Particulate fluxes and transports on the slope between the southern East China Sea and the South Okinawa Trough. Cont Shelf Res.20, 571–597. 10.1016/S0278-4343(99)00086-2 (2000). [Google Scholar]
  • 5.Hung, C. C. et al. Using rare earth elements to constrain particulate organic carbon flux in the East China Sea. Sci. Rep.6, 33880. 10.1038/srep33880 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Kim, M., Hwang, J., Eglinton, T. I., Druffel, E. R. Lateral particle supply as a key vector in the oceanic carbon cycle. Glob. Biogeochem. Cycles34, e2020GB006544 (2020). 10.1029/2020GB006544.
  • 7.Kim, H. J. et al. Influence of atmospheric dust deposition on sinking particle flux in the northwest Pacific. Front. Mar. Sci.10, 1180480. 10.3389/fmars.2023.1180480 (2023). [Google Scholar]
  • 8.Chen, K. S. et al. Enhanced POC export in the oligotrophic northwest Pacific Ocean after extreme weather events. Geophys. Res. Lett.40, 5728–5734. 10.1002/2013GL058300 (2013). [Google Scholar]
  • 9.Hung, C. C. et al. The effect of typhoon on particulate organic carbon flux in the southern East China Sea. Biogeosciences7, 3007–3018. 10.5194/bg-7-3007-2010 (2010). [Google Scholar]
  • 10.Wang, L., Gula, J., Collin, J. & Mémery, L. Effects of mesoscale dynamics on the path of fast-sinking particles to the deep ocean: A modeling study. J. Geophys. Res. Oceans127, e2022JC018799 (2022). 10.1029/2022JC018799.
  • 11.Kwon, E. Y., Primeau, F. & Sarmiento, J. L. The impact of remineralization depth on the air–sea carbon balance. Nat. Geosci.2, 630–635. 10.1038/ngeo612 (2009). [Google Scholar]
  • 12.Lau, N. C., Leetmaa, A., Nath, M. J. & Wang, H. L. Influences of ENSO-induced Indo–western Pacific SST anomalies on extratropical atmospheric variability during the boreal summer. J. Clim.18, 2922–2942 (2005). [Google Scholar]
  • 13.An, S. I. & Wang, B. Mechanisms of locking of the El Niño and La Niña mature phases to boreal winter. J. Clim.14, 2164–2176. 10.1175/1520-0442(2001)014%3c2164:MOLOTE%3e2.0.CO;2 (2001). [Google Scholar]
  • 14.Xie, S. P. et al. Indian Ocean capacitor effect on Indo-Western Pacific climate during the summer following El Niño. J. Clim.22, 730–747 (2009). [Google Scholar]
  • 15.Yeh, S. W. et al. El Niño in a changing climate. Nature461, 511–514. 10.1038/nature08316 (2009). [DOI] [PubMed] [Google Scholar]
  • 16.Ronghui, H. & Yifang, W. The influence of ENSO on the summer climate change in China and its mechanism. Adv. Atmos. Sci.6, 21–32. 10.1007/BF02656915 (1989). [Google Scholar]
  • 17.Wu, B., Zhou, T. & Li, T. Seasonally evolving dominant interannual variability modes of East Asian climate. J. Clim.22, 2992–3005 (2009). [Google Scholar]
  • 18.Cao, Q. et al. Impacts of various types of El Niño-Southern Oscillation (ENSO) and ENSO Modoki on the rainy season over the Huaihe River basin. Int. J. Climatol.39, 2811–2824. 10.1002/joc.6002 (2019). [Google Scholar]
  • 19.Li, G., Chen, L. & Lu, B. A physics-based empirical model for the seasonal prediction of the central China July precipitation. Geophys. Res. Lett.50, e2022GL101463 (2023). 10.1029/2022GL101463
  • 20.Lin, X., Lu, B., Li, G., Gao, C. & Chen, L. Asymmetric impacts of El Niño-Southern Oscillation on the winter precipitation over South China: The role of the India-Burma Trough. Clim. Dyn.19, 1–7. 10.1007/s00382-023-06675-0 (2023). [Google Scholar]
  • 21.Chen, L. et al. Interdecadal change in the influence of El Niño in the developing stage on the central China summer precipitation. Clim. Dyn.59, 1265–1282. 10.1007/s00382-021-06036-9 (2022). [Google Scholar]
  • 22.Chen, L. & Li, G. Interdecadal change in the relationship between El Niño in the decaying stage and the central China summer precipitation. Clim. Dyn.59, 1981–1996. 10.1007/s00382-022-06192-6 (2022). [Google Scholar]
  • 23.Honjo, S., Connell, J. F. & Sachs, P. L. Deep-ocean sediment trap; design and function of PARFLUX Mark II. Deep Sea Res. A.27, 745–753. 10.1016/0198-0149(80)90026-6 (1980). [Google Scholar]
  • 24.Zhang, J., et al. Biological carbon pump responses to multiscale physical processes: A review of sediment trap studies in the South China Sea. Front. Mar. Sci.10 (2023). 10.3389/fmars.2023.1209686.
  • 25.Lee, D. et al. Estimation of the particulate organic carbon to chlorophyll-a ratio using MODIS-Aqua in the East/Japan Sea, South Korea. Remote Sens.12, 840. 10.3390/rs12050840 (2020). [Google Scholar]
  • 26.Yang, W. et al. Utilization of Soot and 210 Po-210 Pb disequilibria to constrain particulate organic carbon fluxes in the northeastern South China Sea. Front. Mar. Sci.8, 694428. 10.3389/fmars.2021.694428 (2021). [Google Scholar]
  • 27.Huang, C. et al. External forcing mechanisms controlling the Qiongdong upwelling in the northern South China Sea during the Holocene. Glob. Planet. Change220, 104021. 10.1016/j.gloplacha.2022.104021 (2023). [Google Scholar]
  • 28.Kim, M. et al. Carbon cycling in the East Sea (Japan Sea): A review. Front. Mar. Sci.9, 938935. 10.3389/fmars.2022.938935 (2022). [Google Scholar]
  • 29.Kim, M. et al. Biogeochemical properties of sinking particles in the southwestern part of the East Sea (Japan Sea). J. Mar. Syst.167, 33–42. 10.1016/j.jmarsys.2016.11.001 (2017). [Google Scholar]
  • 30.Kim, M. et al. Influence of sediment resuspension on the biological pump of the southwestern East Sea (Japan Sea). Front. Earth Sci.8, 538412. 10.3389/feart.2020.00144 (2020). [Google Scholar]
  • 31.Kim, M. et al. Spatiotemporal variation of sterols in sediment as markers of primary production and ocean sewage dumping in the southwestern East Sea (Japan Sea). Mar. Pollut. Bull.205, 116680. 10.1016/j.marpolbul.2024.116680 (2024). [DOI] [PubMed] [Google Scholar]
  • 32.Hidayat, M. N., Wafdan, R., Ramli, M., Muchlisin, Z.A. & Rizal, S. Relationship between chlorophyll-a, sea surface temperature, and sea surface salinity. Glob. J. Environ. Sci. Manag.9, 389–402 (2023). 10.22034/gjesm.2023.03.03.
  • 33.Gaube, P., Chelton, D. B., Strutton, P. G. & Behrenfeld, M. J. Satellite observations of chlorophyll, phytoplankton biomass, and Ekman pumping in nonlinear mesoscale eddies. J. Geophys. Res. Oceans118, 6349–6370. 10.1002/2013JC009027 (2013). [Google Scholar]
  • 34.Dufois, F. et al. Impact of eddies on surface chlorophyll in the South Indian Ocean. J. Geophys. Res. Oceans119, 8061–8077. 10.1002/2014JC010164 (2014). [Google Scholar]
  • 35.Honjo, S., Dymond, J., Collier, R. & Manganini, S. J. Export production of particles to the interior of the equatorial Pacific Ocean during the 1992 EqPac experiment. Deep Sea Res. II. Honjo42, 831–870. 10.1016/0967-0645(95)00034-N (1995). [Google Scholar]
  • 36.Kim, H. J. et al. Influence of Asian monsoon and ENSO events on particle fluxes in the western subtropical Pacific. Deep Sea Res.I(90), 139–151. 10.1016/j.dsr.2014.05.002 (2014). [Google Scholar]
  • 37.Kim, H. J. et al. Suppressed CO2 outgassing by an enhanced biological pump in the Eastern Tropical Pacific. JGR Oceans124, 7962–7973. 10.1029/2019JC015287 (2019). [Google Scholar]
  • 38.Behrenfeld, M. J. & Falkowski, P. G. Photosynthetic rates derived from satellite-based chlorophyll concentration. Limnol. Oceanogr.42, 1–10. 10.4319/lo.1997.42.1.0001 (1997). [Google Scholar]
  • 39.Park, H., Kim, G., Lee, J. S., Kim, S. H. Benthic fluxes of total dissolved amino acids to the ocean: Results from the East/Japan Sea. In Goldschmidt 2023 Conference. 10.1002/lol2.10363 (2023).
  • 40.Yoo, S. & Park, J. Why is the southwest the most productive region of the East Sea/Sea of Japan?. J. Mar. Syst.78, 301–315. 10.1016/j.jmarsys.2009.02.014 (2009). [Google Scholar]
  • 41.MODIS. NASA Goddard Space Flight Center, Ocean Ecology Laboratory, Ocean Biology Processing Group, 2018. Moderate-Resolution Imaging Spectroradiometer (MODIS) Aqua Chlorophyll Data (2018).
  • 42.Joo, H. et al. Long-term annual primary production in the Ulleung Basin as a biological hot spot in the East/Japan Sea. J. Geophys. Res. Oceans119, 3002–3011. 10.1002/2014JC009862 (2014). [Google Scholar]
  • 43.Kwak, J. H. et al. Monthly measured primary and new productivities in the Ulleung Basin as a biological" hot spot" in the East/Japan Sea. Biogeosciences10, 4405–4417. 10.5194/bg-10-4405-2013 (2013). [Google Scholar]
  • 44.CPC. ONI data in Climate Prediction Center (National Weather Service). https://origin.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/ONI_v5.php (Accessed 23 Sep 2023) (2023).
  • 45.Copernicus. Oceanographic data (SSH, SST, SSS, VEL) in Copernicus Marine Service. https://data.marine.copernicus.eu/product/GLOBAL_REANALYSIS_PHY_001_031/services (2023) (Accessed 23 Sep 2023).
  • 46.CLS. Mesoscale Eddy Trajectories Atlas META3.1exp Delayed-Time All Satellites (3.1exp DT Allsat) [Data Set] (CNES, 2021). 10.24400/527896/A01-2021.001.
  • 47.Lee, C. W., Tseng, Y. H., Sui, C. H., Zheng, F., Wu, E. T. Characteristics of the prolonged El Niño events during 1960–2020. Geophys. Res. Lett. Lee47, e2020GL088345 (2020). 10.1029/2020GL088345.
  • 48.Jin, H., Park, Y. G., Pak, G. & Kim, Y. H. On the persistence of warm eddies in the East Sea. Sea J. Korean Soc. Oceanogr.24, 318–331 (2019). [Google Scholar]
  • 49.Yun, J. Y. et al. ENSO teleconnection to the isopycnal depth fluctuations of the East/Japan Sea Intermediate Water in the Ulleung Basin during 1968–2002. J. Phys. Oceanogr.46, 2675–2694. 10.1175/JPO-D-15-0225.1 (2016). [Google Scholar]
  • 50.Shin, H. R. et al. Physical characteristics and classification of the Ulleung Warm Eddy in the East Sea (Japan Sea). Sea24, 298–317. 10.7850/jkso.2019.24.2.298 (2019). [Google Scholar]
  • 51.Dunstan, P. K. et al. Global patterns of change and variation in sea surface temperature and chlorophyll a. Sci. Rep.8, 14624. 10.1038/s41598-018-33057-y (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Cael, B. B., Bisson, K. & Follett, C. L. Can rates of ocean primary production and biological carbon export be related through their probability distributions?. Glob. Biogeochem. Cycles32, 954–970. 10.1029/2017GB005797 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Cael, B. B., et al. Open ocean particle flux variability from surface to seafloor. Geophys. Res. Lett.48, e2021GL092895 (2021). 10.1029/2021GL092895.
  • 54.Banse, K. New views on the degradation and disposition of organic particles as collected by sediment traps in the open sea. Deep Sea Res. A37, 1177–1195. 10.1016/0198-0149(90)90058-4 (1990). [Google Scholar]
  • 55.Armstrong, R. A., Lee, C., Hedges, J. I., Honjo, S. & Wakeham, S. G. A new, mechanistic model for organic carbon fluxes in the ocean based on the quantitative association of POC with ballast minerals. Deep Sea Res. II. Armstrong49, 219–236. 10.1016/S0967-0645(01)00101-1 (2001). [Google Scholar]
  • 56.Suess, E. Particulate organic carbon flux in the oceans—Surface productivity and oxygen utilization. Nature288, 260–263. 10.1038/288260a0 (1980). [Google Scholar]
  • 57.Belcher, A. et al. Depth-resolved particle-associated microbial respiration in the northeast-Atlantic. Biogeosci. Belcher.13, 4927–4943. 10.5194/bg-13-4927-2016 (2016). [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material 1 (211.3KB, docx)

Data Availability Statement

All ocean sinking particle data used for this manuscript are available in the article/Supplementary Material of Kim et al. (2020; https://www.frontiersin.org/articles/10.3389/feart.2020.00144/full#supplementary-material). In the present study, we described the data availability in the Regional setting session. All the data we analyzed are summarized in Table 1, S1, and S2.

Articles from Scientific Reports are provided here courtesy of Nature Publishing Group

RESOURCES