Next Article in Journal
Using Multi-Source Geospatial Information to Reduce the Saturation Problem of DMSP/OLS Nighttime Light Data
Previous Article in Journal
Calibration and Validation of CYGNSS Reflectivity through Wetlands’ and Deserts’ Dielectric Permittivity
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Remote Sensing
Volume 14
Issue 14
10.3390/rs14143263
remotesensing-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Phytoplankton Blooms Expanding Further Than Previously Thought in the Ross Sea: A Remote Sensing Perspective

by
Shuangling Chen
* and
Yu Meng
State Key Laboratory of Satellite Ocean Environment Dynamics, Second Institute of Oceanography, Ministry of Natural Resources, Hangzhou 310012, China
*
Author to whom correspondence should be addressed.
Remote Sens. 2022, 14(14), 3263; https://doi.org/10.3390/rs14143263
Submission received: 19 May 2022 / Revised: 25 June 2022 / Accepted: 2 July 2022 / Published: 6 July 2022
(This article belongs to the Section Ocean Remote Sensing)
Browse Figures
Versions Notes

Abstract

:
Accurate and robust measurements from ocean color satellites are critical to studying spatial and temporal changes of surface ocean properties. Satellite-derived Chlorophyll-a (Chl) is an important parameter to monitor phytoplankton blooms on synoptical scales, particularly in remote seas. However, the present NASA standard Chl algorithm tends to strongly underestimate the Chl in the Ross Sea. Based on a locally-tuned Chl algorithm in the Ross Sea and using the data record from MODIS between 2002 and 2020, here we investigated the spatial expansion of phytoplankton blooms in the Ross Sea. Our results show the geometric areas of the phytoplankton blooms could reach (7.20 ± 2.8) × 104 km2 on average, which was ~3.1 times that of those identified using the NASA default Chl algorithm. Spatially, blooms were frequently identified on the shelf of the Ross Sea polynya with a typical chance of ≥80%. In the context of climate change and global warming, the general decrease and interannual dynamics of sea ice cover tends to affect solar light penetration and surface seawater temperature, which were found to regulate the spatial expansion of the phytoplankton blooms over the years. Statistical analyses showed that the spatial coverages of the phytoplankton blooms were significantly correlated with sea surface temperature (Spearman correlation coefficient R = 0.55, at p < 0.05), sea surface wind speed (R = 0.42, at p < 0.05), and sea ice concentration (R = −0.84, at p < 0.05), yet without significant long-term (>10 years) trends over the study period. The stronger phytoplankton blooms than those previously observed may indicate larger carbon sequestration, which needs to be investigated in the future. More valid satellite observations under cloud covers will further constrain the estimates.

1. Introduction

Since the start of industrialization and global warming, more than 40% of the CO2 and 75% of the heat that has been produced anthropogenically has been absorbed by the Southern Ocean [1]. The Ross Sea is the most productive marginal sea of the Southern Ocean in austral spring and summer when the coastal polynyas form and phytoplankton blooms. Phytoplankton blooms can efficiently absorb CO2 in the upper ocean, and they are important to alleviate ocean acidification and sequence ocean carbon [2,3,4,5]. The Ross Sea plays a critical role in marine biogeochemical cycles and air-sea CO2 fluxes on global scales [6,7]. Studies show that the Ross Sea continental shelf is highly productive with an annual primary productivity of 23.4 Tg C yr−1, which accounts for over a third of the total productivity of the Antarctic shelf [6,8]. The high rates of primary productivity during austral spring and summer result in low surface water pCO2 [9], which drives the Ross Sea to be a strong CO2 sink of 7.5 Tg C yr−1, and this carbon sink is ~27% of the total CO2 taken up by the entire Southern Ocean [10,11,12]. Therefore, better quantification of the phytoplankton bloom dynamics would improve our understanding of the biogeochemical and carbon cycling in this important region.
The quantity and quality of phytoplankton are affected by various environmental conditions, such as light conditions, sea temperature, nutrients, and phytoplankton species composition [13,14]. There are two dominant phytoplankton function groups in the Ross Sea, including haptophyte P. antarctica and diatoms [15,16]. The phytoplankton growth in the Ross Sea is believed to be seasonally limited, first in austral spring by irradiance, and then in summer by biologically available iron [17]. During austral spring and early summer (November–December), the phytoplankton blooms are dominated by P. antarctica, which accounts for ~95% of the total phytoplankton blooms in the polynya [18]. Smith & Kaufman [19] found that the particulate organic carbon (POC, mg m−3) covaried with Chlorophyll-a concentrations (Chl, mg m−3) during this time period, suggesting the Chl dynamics are mainly controlled by phytoplankton biomass [20]. In late December and early January, the abundance of P. antarctica rapidly declines, yet diatoms become more prevalent, most likely due to their adaptation to the extreme iron limitation in the polynya [18]. During this period, POC decoupled with Chl experiences a notable increase largely independent to Chl changes [19].
As a result of the stresses of global climate change, the physical conditions of the Ross Sea have been changing in recent decades. Specifically, associated with the temperature increase in the atmosphere [21], the average sea ice extension and duration are decreasing in the Ross Sea polynya [6], and the mixed layer depths are getting shallower [22]. The phytoplankton dynamics are expected to change in response, particularly on interannual time scales. Phytoplankton blooms occur when the phytoplankton biomass shows a sharp increase within a short time period. Chl is an important indicator for biomass and phytoplankton abundance in the ocean ecosystem [23,24]. The Chl dynamics in the Ross Sea in recent decades can be used to examine the long-term changing patterns of phytoplankton blooms in this area. However, field research expeditions are very limited due to the remote and harsh environment in the Ross Sea [17,25]. Based on a one-dimensional biogeochemical modeling, Kaufman et al. [26] revealed that the phytoplankton primary production in the Ross Sea would increase 5% by midcentury, yet variabilities in boundary conditions, particularly due to climate change, may bring large uncertainties to the model projections.
In contrast, ocean color remote sensing has great advantages in monitoring the dynamics of phytoplankton blooms. The accumulation of ocean color satellite remote sensing of Chl in recent decades provides a reliable data source, which serves an alternative way for studying the phytoplankton bloom dynamics on synoptical time scales [27]. A few studies have attempted to examine the seasonal-interannual variabilities of Chl and/or fluorescence in the Ross Sea [28,29], however, they most focused on short-term analyses (≤4-year period). Using a 15-years (2002–2017) data record from MODIS, Park et al. [30] found that the amplitude of phytoplankton blooms in the Ross Sea tended to increase, which was attributed to the formation of open water area. However, study shows that the MODIS standard Chl data products were strongly underestimated in the Ross Sea [31] (see more details in the next paragraph). As a result, there are still lots of uncertainties in the long-term changing patterns of the phytoplankton blooms. More importantly, the spatial expansion dynamics of the phytoplankton blooms in recent decades is generally unknown.
Indeed, accuracy is the priority in their spatial and temporal analyses of the ocean color remote sensing data. It is known that the bio-optical properties in the polar regions are significantly different from those in the mid- and low-latitude oceans [32]. In polar oceans, cloud covers are typically persistent and solar angles are often small even in summer. To adapt to the low light conditions in polar oceans, the phytoplankton pigment packaging effects are stronger, resulting in a lower light absorption per Chl [33,34,35]. This raises the following question: is the NASA standard Chl algorithm for the global ocean applicable for the Ross Sea? Chen et al. [31] recently evaluated the accuracy of the NASA standard Chl algorithm based on an extensive cruise dataset in the history in the Ross Sea, and found that the Chl was strongly underestimated by the NASA standard algorithm with a root mean square difference (RMSD) of 4.72 mg m−3 and mean bias of −3.48 mg m−3. As such, we formulated our hypothesis that the use of NASA standard Chl would also significantly underestimate the strength and the spatial expansion of phytoplankton blooms in the Ross Sea. To verify this hypothesis and fill in the knowledge gap of the interannual changes of the spatial expansion of phytoplankton blooms in recent decades, we chose to use the locally-tuned Chl algorithm proposed by Chen et al. [31]. The objectives of this study include: (1) to use the new Chl data products to establish a complete data record of phytoplankton bloom dynamics in the Ross Sea, (2) to investigate their spatial/temporal variations as well as their possible long-term trends in recent decades, and (3) to understand their driving mechanisms under climate change.

2. Data and Methods

2.1. Data

The NASA standard daily Level-3 data products (version R2018.0, Goddard Space Flight Center (GSFC), Greenbelt, Maryland, USA) derived from measurements by the Moderate Resolution Imaging Spectroradiometer (MODIS) on the Aqua satellite were used in this study. Data of remote sensing reflectance (Rrs) at a band of 667 nm and the NASA standard data products of Chl between 2002 and 2020 with a spatial resolution of ~4 km, were downloaded from the NASA GSFC (https://oceancolor.gsfc.nasa.gov/, accessed on 9 January 2019). Meanwhile, the daily sea surface temperature (SST, °C) and photosynthetic active radiation (PAR, Einstein m−2 d−1) from MODIS were also obtained from the GSFC. The Cross-Calibrated Multi-Platform (CCMP, V2.0, Remote Sensing Systems, Santa Rosa, CA, USA) gridded (0.25° × 0.25°) Level-3 daily surface vector winds, including U and V wind components, were downloaded from the Remote Sensing Systems (https://www.remss.com/measurements/ccmp/, accessed on 16 April 2022). The corresponding sea surface wind speed (SSW, m/s) and wind direction (Wdir, in range of [−180°, 180°]) were calculated from the U and V vectors. The satellite measurements of daily sea ice concentration (SIC, %) were derived from the Scanning Multichannel Microwave Radiometer (SMMR) on the Nimbus-7 satellite, the Special Sensor Microwave/Imager (SSM/I) and the Special Sensor Microwave Imager/Sounder (SSMIS) sensors on the Defense Meteorological Satellite Program’s (DMSP) satellites. The daily SIC data between 2002 and 2020, with a spatial resolution of 25 km, were obtained from the National Snow & Ice Data Center (https://nsidc.org/data/nsidc-0079, accessed on 11 April 2022).

2.2. Methods

To investigate the expansion dynamics of phytoplankton blooms in the Ross Sea, we first derived the locally-tuned Chl data products from MODIS using the Chl algorithm developed in a recent study [31]. This algorithm was locally tuned for the Ross Sea based on the remote sensing reflectance at band of 667 nm (Rrs667) (Equation (1)), which was proved to outperform both the band-ratio-based OCx algorithms and band-difference-based OCI algorithms commonly used in the NASA standard data processing routine [36,37,38]. One possible reason for the failure of the NASA standard Chl algorithm in the Ross Sea is the distinctiveness of the bio-optical properties in the polar marginal seas, where the solar angles are typically small even in austral summer, and phytoplankton acclimates to the low light conditions. More details can be found in Chen et al. [31]. We applied Equation (1) to all the MODIS daily Rrs667 data between 2002 and 2020 to obtain the locally-tuned daily Chl images. Note that considering the dense ice coverages between March and October, most valid observations are only permitted in austral summer (November-February), within restricted areas, thus only data in this time window were processed in this study, and a complete bloom season was defined from November to the following February in the next year [30,39].
Chl = 12352.0 × Rrs667 − 2.8477
However, even for the daily images in the chosen time window, tremendous data gaps still exist due to severe cloud contamination, making it difficult to characterize the spatial distribution of phytoplankton dynamics in the Ross Sea. Alternatively, to increase the spatial coverages of cloud-free data, we generated weekly Chl composites from the daily Chl images and used the weekly composites to identify phytoplankton blooms. A phytoplankton bloom was defined with a Chl threshold of 2.0 mg m−3 [30]. Following this definition, patches with Chl greater than this threshold were flagged to indicate potential algal bloom signals. In this way, phytoplankton bloom masks were created for each weekly Chl image. Meanwhile, to illustrate the difference between bloom masks identified using either the locally-tuned Chl or the NASA standard Chl, the phytoplankton bloom masks were also produced for the NASA standard weekly Chl images following the same processing procedure described above. It should be clarified that, all the weekly Chl images were projected using the Lambert projection to maintain the real shapes of the geometrically small areas in polar regions.
Based on the bloom masks identified from the weekly Chl images, we derived the bloom masks for each year between 2002 and 2020. Then the geometric bloom areas in each year were calculated based on the bloom pixels in the bloom masks. As noted above, although the MODIS Level-3 data were gridded at a spatial resolution of 4 km, the spatial area of a pixel in the Ross Sea is much smaller than 4 km × 4 km because the pixel sizes geometrically become smaller poleward, and this effect was fully considered in our geometric area calculation based on Equation (2). Furthermore, to derive the phytoplankton bloom frequency map in the Ross Sea, for each pixel on annual scales, we counted the total number of blooms observed over the study period (i.e., 19 years), and then calculated the bloom frequency by dividing the total number of bloom observations by the total number of valid satellite observations. The bloom frequency was regarded as the possibility of phytoplankton bloom occurrence in the study area.
Si = (π/180.0) × r2 × |sin(latiupper/180.0 × π) − sin(latibottom/180.0 × π)| × |loniwest − lonieast|
where Si represents the geometric area of pixel i, r is the mean radius of the Earth (r = 6378.0 km), latiupper and latibottom are the upper and bottom latitude bounds of the pixel, and loniwest and lonieast are the longitude bounds of the pixel.

3. Results & Discussion

3.1. Expansion and Frequency of Phytoplankton Blooms

Figure 1a shows the weekly composite of Chl image from 12–18 February 2005 using the locally-tuned Chl algorithm. Distinct phytoplankton blooms were observed on the coast of Victoria Land and in the open ocean waters of northern Ross Sea, with Chl reaching ~15.0 mg m−3 in the Terra Nova Bay. Specifically, the phytoplankton bloom in the open ocean waters was huge and extended into large areas over the space, which was clearly presented in the corresponding bloom mask in Figure 1b. It is noted that in the northeast Ross Sea, sparse Chl were observed due to strong cloud/ice contamination, and these Chl data were noisy around the cloud/ice edges, with some greater than 2 mg m−3 and identified as false bloom by accident. In contrast, Figure 1c,d shows the results based on the NASA standard Chl algorithm. Clearly, the values of Chl were significantly underestimated comparing to those derived from the locally-tuned algorithm (Figure 1a). As a result, only a tiny patch of relatively high Chl (i.e., 2.0–2.5 mg m−3) in the open ocean waters were identified as phytoplankton bloom, and the overall spatial expansion and strength of the phytoplankton blooms were strongly dampened and much smaller than those captured by the locally-tuned Chl, particularly in the open ocean waters of the Ross Sea.
We applied the Rrs667-based Chl algorithm to all the MODIS images between 2002 and 2020, and generated a map of phytoplankton bloom frequency (Figure 2) that occurred in the past 19 years following the procedures described in Section 2.2. The bloom frequency represents the chances of phytoplankton bloom to occur in the Ross Sea from a statistical perspective. It is found that there are generally high frequencies (≥80%) of blooms shown on the continental shelf of Ross Sea, and the bloom frequency in the Terra Nova Bay could reach even 100%, indicating the popularity and persistency of phytoplankton blooms in these regions every year. The overall spatial distributions of the bloom areas identified were consistent but extended much wider than those shown in studies based on the NASA standard Chl algorithm [17,30], suggesting that the spatial expansion of phytoplankton blooms in the Ross Sea is much larger than previously thought, and it also verified our hypothesis (see the Introduction section). Specifically, the phytoplankton blooms had a much wider expansion off the coast of Victoria Land to the offshore, but kept narrow on the eastern Ross Ice Shelf. Based on the locally-tuned Chl, blooms in the open ocean waters were also found from time to time with a chance of <20%, suggesting the rareness of phytoplankton blooms in open ocean waters. However, caution is needed for the interpretation of the bloom frequency in the open ocean waters. On one hand, strong cloud contaminations are persistent in these waters due to frequent thick cloud covers, as such, the data noises in Chl (e.g., Figure 1a) could result in some false blooms in these open waters (Figure 2). On the other hand, the Rrs667-based Chl algorithm was tuned based on observations concentrating on the continental shelf of the Ross Sea, because of the empirical characteristic of the locally-tuned algorithm (Equation (1)), it may not work well beyond the shelf waters despite the limited validation approved its general applicability [31]. Therefore, uncertainties may exist in the derived bloom frequency in these open ocean waters. In addition, it is noted that along the edges of high bloom frequencies on the continental shelf towards the coastline, there is a narrow belt where a lower chance of phytoplankton blooms is detected.
It is known that satellite observations in the Ross Sea suffer from persistent cloud covers even in austral spring and summer [40]. To exclude the possibility that high frequencies of blooms were artificially introduced when the only a few times of valid satellite observations just perfectly captured bloom signals by accident, we generated the observation frequency map with valid satellite observations in the period of 2002–2020 (Figure 3a). It is found that for all the areas with high frequencies of blooms identified, the corresponding frequency of valid satellite observations were all high, which suggests the statistics of high bloom frequencies were based on sufficient satellite data observations. As such, we are confident that the high bloom frequencies shown in Figure 2 should be realistic. Besides, it is noteworthy that areas with high observing frequencies do not necessarily associate with high chances of phytoplankton blooms. For example, the areas of high bloom frequencies only expanded to ~74°S northward on the western shelf (Figure 2), yet valid satellite observations are always available for areas extending to ~71°S (Figure 3a). On the east shelf of the Ross Sea, areas with high observation frequencies had a wider spatial coverage but with a narrow range of bloom popularity (Figure 2). Moreover, corresponding to the narrow belt of low bloom frequencies along the coastline, the valid satellite observations were also limited, which seems to be caused by land and/or ice contamination along the coastline. For such cases, these low bloom frequencies identified could be artificial as well due to the limited data observations, particularly considering the great expansion of high bloom frequencies on the adjacent continental shelf. Still, more valid satellite observations under cloud covers will further constrain the estimates.
Following the procedures described in Section 2.2, we also generated the phytoplankton bloom frequency using the NASA standard Chl (Figure 3b). It is seen that relatively high bloom frequencies (≥40%) only appear on the west shelf (southward of 75°S, and to the west of 168°E). More importantly, although the Terra Nova Bay and the east shelf are typical areas with phytoplankton blooms identified from in situ measurements [15,16,18], they were rarely captured from satellites using the NASA standard Chl processing routine, which again proves the strong underestimation of phytoplankton bloom extension from ocean color using the NASA default algorithm.

3.2. Interannual Dynamics of Phytoplankton Blooms

With the spatial expansion of phytoplankton blooms identified, we quantified the geometric areas of the blooms for each year between 2002 and 2020 (Figure 4a). In general, the bloom areas identified using the locally-tuned Chl could extend to (7.20 ± 2.86) × 104 km2 on average, which is ~3.1 times of that identified using the NASA default Chl, again, statistically proving our hypothesis that the spatial expansion of phytoplankton blooms was strongly underestimated using the NASA standard default Chl algorithm. The bloom areas varied on interannual time scales ranging between 1.38 × 104 km2 and 12.58 × 104 km2. Specifically, a short-term increase and decrease in bloom areas was observed in the period of 2003–2006 and 2006–2008, respectively, with a maximum in 2006 and small-scale interannual dynamics in the following years (i.e., after the year of 2008). In contrast, the bloom areas identified using the NASA standard Chl were much smaller, but the interannual variation patterns were similar with a strong interannual correlation with the bloom areas identified based on the locally-tuned Chl, and the Spearman correlation coefficient R was 0.90 at significant level (p < 0.05).
We also investigated the corresponding phytoplankton bloom strengths in terms of mean Chl within the bloom areas (Figure 4b). On average, the phytoplankton blooms identified using the locally-tuned Chl had a mean Chl of 4.28 ± 0.38 mg m−3, which was stronger than that using the NASA standard Chl (i.e., bloom strength of 3.52 ± 0.59 mg m−3 on average). On interannual time scales, the bloom strengths showed slight variations over the years, yet without any interannual trend. Comparing to Figure 4a, it is found that larger bloom areas were not necessarily associated with higher bloom strengths, which seems reasonable because the bloom areas were identified as long as the Chl of a pixel was ≥2.0 mg m−3 while the bloom strengths were derived from the averages of Chl within the bloom areas. The bloom strengths characterized using the locally-tuned Chl were generally stronger than those using the NASA standard Chl both on average and on interannual time scales, with relatively small difference varying between 0.1 and 2.0 mg m−3. It suggests that despite the big underestimation of Chl using the NASA standard processing routine, the spatial-averaged bloom strengths still kept at similar but lower level as those characterized by the locally-tuned Chl algorithm. In addition, we also compared the interannual Chl dynamics in the study region (i.e., including both the bloom and non-bloom areas) (Figure 4c), which indicates the mean state of phytoplankton biomass. Statistically, the NASA standard Chl was on level of 1.00 ± 0.33 mg m−3 between 2002 and 2020, which was much smaller than the Chl (an average of 3.41 ± 1.18 mg m−3) based on the locally-tuned algorithm, consistent with the results shown in Chen et al. [31]. More importantly, the locally-tuned high Chl from ocean color seems more reasonable to support the high primary productivity [6,8] and high CO2 uptake [10,11] reported based on in situ measurements in the Ross Sea.
To further understand the interannual variabilities of the phytoplankton blooms (Figure 4), time series data of SST, PAR, SSW, Wdir, and SIC, were processed for the study time period and presented in Figure 5. It is noted that the expansion of phytoplankton blooms was generally larger at higher SST, and vice versa, suggesting that phytoplankton blooms and expands more easily in warm waters. Indeed, phytoplankton growth and metabolism are regulated by seawater temperature and favored with warm temperature [41,42]. Statistically, the spatial expansion of phytoplankton blooms had a strong positive correlation with SST (R = 0.55, at p < 0.05) and SSW (R = 0.42, at p < 0.1) at significant level. Both SST and SSW constrains stratification and mixed layer formation, providing the phytoplankton blooms of P. antarctica and diatoms with sufficient nutrients and light [43]. It is known that sea ice is an important factor restricting solar penetration and thus warming of surface layers [44]. Here we found that the SIC had a strong negative correlation with PAR (R = −0.55, at p < 0.05) and SST (R = −0.61, at p < 0.05) at significant level. In austral spring and summer, the SIC kept at a low level of 11.1 ± 2.4% (Figure 5e), and the wind blows in a northeasterly direction (Wdir of 0–50°, Figure 5d), favoring the formation of polynyas in the Ross Sea [30]. As such, we found a strong negative correlation between SIC and the geometric areas of phytoplankton blooms with R of −0.84 at significant level (p < 0.05), suggesting the interannual dynamics of the spatial expansion of phytoplankton blooms were mainly driven by SIC. The interannual changes of the SIC in the Ross Sea were reported to modulated by the El Nino Southern Oscillation (ENSO) [6,45,46]. In the context of global warming and climate change, the SIC showed a general decreasing trend with a distinct short-term decrease and increase over the years. For example, 2003–2006 and 2006–2008, the SIC showed a decreasing and increasing pattern, respectively (Figure 5e); the decrease of SIC favored solar light penetration (i.e., increasing PAR) and surface warming of seawater (i.e., increasing SST), and vice versa. Correspondingly, there was an obvious short-term increase and decrease in the bloom areas (Figure 4a). Particularly, in the bloom season of 2002–2003, the SIC reached maximum over the whole study period, and the corresponding SST and PAR were abnormally low, as a result, the phytoplankton blooms showed the most limited spatial expansion in response. In contrast, in the bloom season of 2005–2006, the SIC reached minimum, which favored the solar light penetration and surface warming, and the phytoplankton blooms exhibited the largest spatial expansion. In recent years (i.e., 2019–2021), the SIC kept at pretty low level at 9.75~11.35%, the SST presented a steady increase, and the phytoplankton tended to bloom over a wider area.
In fact, the complex nonlinearities between the environmental forcings and phytoplankton productivity co-determine the phytoplankton dynamics [19]. As such, despite the dominant role of climate change in SIC, the complex interactions of different physical forcings (e.g., surface warming, wind speed, wind direction, vertical mixing), which modulates the light and nutrient environment as well as seawater temperature, should also contribute to regulating the spatial expansion of phytoplankton blooms in the Ross Sea. Specifically, due to the large solar angles, the satellite-observed PAR was quite low with an average of 26.1 ± 1.7 Einstein m−2 d−1 even in austral spring and summer, and the pigments of phytoplankton, therefore, tended to package stronger to adapt to the low light conditions and to support the phytoplankton growth [17,35]. We speculate that the environmental forcings were nonlinearly correlated with dynamics of phytoplankton bloom strengths and spatial expansion in a more complex manner.

4. Conclusions

Accurate estimations of surface Chl from ocean color is critical for monitoring phytoplankton blooms on synoptical scales. Using the locally-tuned Chl algorithm developed in a recent study for the Ross Sea and a data record from MODIS between 2002 and 2020, we found that the phytoplankton blooms in the Ross Sea expanded much wider than thought. The geometric bloom areas identified were significantly larger and the bloom observing frequencies were significantly higher than those based on the NASA standard algorithm. The interannual dynamics of the spatial expansion of the phytoplankton blooms were mainly driven by sea ice coverages under climate change, modulated together with other physical forcings. The phytoplankton blooms were stronger than previously thought, and they, therefore, play an even more important role in regional carbon export and air-sea CO2 fluxes, and this needs to be investigated in future studies.

Author Contributions

Conceptualization, S.C.; Formal analysis, Y.M.; Funding acquisition, S.C.; Methodology, S.C.; Project administration, S.C.; Visualization, S.C. and Y.M.; Writing—original draft, S.C.; Writing—review & editing, S.C. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Qianjiang Talent Program of Zhejiang Province [QJD2002034], National Natural Science Foundation of China [41906159 and 42030708], and National Key Research and Development Program of China [2021YFE0117600] The APC was funded by National Natural Science Foundation of China [42030708].

Data Availability Statement

The NASA standard Level-3 daily data products including Chl, PAR, Rrs(667) and SST between 2002 and 2020 were obtained the NASA Goddard Space Flight Center (https://oceancolor.gsfc.nasa.gov/, accessed on 9 January 2019), the CCMP surface vector winds data were obtained from the Remote Sensing Systems (https://www.remss.com/measurements/ccmp/, accessed on 16 April 2022), and the SIC data were obtained from the National Snow & Ice Data Center (https://nsidc.org/data/nsidc-0079, accessed on 11 April 2022). More details on the data description of each dataset can be found in Section 2.1 (Data). We state that it is straightforward to generate the locally-tuned daily Chl images using the local-tuned Chl algorithm (Equation (2)).

Acknowledgments

The authors thank the Qianjiang Talent Program of Zhejiang Province, the National Natural Science Foundation of China, and National Key Research and Development Program of China for the funding support of this research. The authors also thank NASA Goddard Space Flight Center (http://oceancolor.gsfc.nasa.gov/, accessed on 9 January 2019), the Remote Sensing Systems (https://www.remss.com/, accessed on 16 April 2022), and the National Snow & Ice Data Center (https://nsidc.org/data/nsidc-0079, accessed on 11 April 2022) for maintaining and providing the satellite data used in this study. The authors are grateful for the constructive comments from the three anonymous reviewers that helped improve the manuscript.

Conflicts of Interest

The authors declare no conflict of interest involved.

References

  1. Frölicher, T.; Sarmiento, J.L.; Paynter, D.J.; Dunne, J.P.; Krasting, J.; Winton, M. Dominance of the Southern Ocean in Anthropogenic Carbon and Heat Uptake in CMIP5 Models. J. Clim. 2015, 28, 862–886. [Google Scholar] [CrossRef]
  2. Wang, S.; Moore, J.K. Variability of primary production and air-sea CO2 flux in the Southern Ocean. Glob. Biogeochem. Cycles 2012, 26, GB1008. [Google Scholar] [CrossRef] [Green Version]
  3. Hu, Y.-B.; Liu, C.-Y.; Yang, G.-P.; Zhang, H.-H. The response of the carbonate system to a green algal bloom during the post-bloom period in the southern Yellow Sea. Cont. Shelf Res. 2015, 94, 1–7. [Google Scholar] [CrossRef]
  4. Chen, S.; Hu, C. Environmental controls of surface water pCO2 in different coastal environments: Observations from marine buoys. Cont. Shelf Res. 2019, 183, 73–86. [Google Scholar] [CrossRef]
  5. Rivaro, P.; Ianni, C.; Raimondi, L.; Manno, C.; Sandrini, S.; Castagno, P.; Falco, P. Analysis of physical and bio-geochemical control mechanisms on summertime surface carbonate system variability in the western Ross Sea (Antarctica) using in situ and satellite data. Remote Sens. 2019, 11, 238. [Google Scholar] [CrossRef] [Green Version]
  6. Schine, C.M.; van Dijken, G.; Arrigo, K.R. Spatial analysis of trends in primary production and relationship with large-scale climate variability in the Ross Sea, Antarctica (1997–2013). J. Geophys. Res. Oceans 2016, 121, 368386. [Google Scholar] [CrossRef] [Green Version]
  7. Deppeler, S.L.; Davidson, A.T. Southern Ocean Phytoplankton in a Changing Climate. Front. Mar. Sci. 2017, 4, 40. [Google Scholar] [CrossRef] [Green Version]
  8. Arrigo, K.R.; Van Dijken, G.; Bushinsky, S. Primary production in the Southern Ocean, 1997–2006. J. Geophys. Res. Earth Surf. 2008, 113, C08004. [Google Scholar] [CrossRef]
  9. Tortell, P.D.; Guéguen, C.; Long, M.C.; Payne, C.D.; Lee, P.; DiTullio, G.R. Spatial variability and temporal dynamics of surface water pCO2, ΔO2/Ar and dimethylsulfide in the Ross Sea, Antarctica. Deep. Sea Res. Part I Oceanogr. Res. Pap. 2011, 58, 241–259. [Google Scholar] [CrossRef]
  10. Arrigo, K.R.; van Dijken, G.; Long, M. Coastal Southern Ocean: A strong anthropogenic CO2 sink. Geophys. Res. Lett. 2008, 35, 21–26. [Google Scholar] [CrossRef]
  11. DeJong, H.B.; Dunbar, R.B. Air-sea CO2 exchange in the Ross Sea, Antarctica. J. Geophys. Res. Oceans 2017, 122, 8167–8181. [Google Scholar] [CrossRef]
  12. Van der Linden, F.C.; Tison, J.; Champenois, W.; Moreau, S.; Carnat, G.; Kotovitch, M.; Fripiat, F.; Deman, F.; Roukaerts, A.; Dehairs, F.; et al. Sea Ice CO2 Dynamics Across Seasons: Impact of Processes at the Interfaces. J. Geophys. Res. Oceans 2020, 125, e2019JC015807. [Google Scholar] [CrossRef]
  13. Morris, I.; Glover, H.E.; Yentsch, C.S. Products of photosynthesis by marine phytoplankton: The effect of environmental factors on the relative rates of protein synthesis. Mar. Biol. 1974, 27, 1–9. [Google Scholar] [CrossRef]
  14. Lee, J.H.; Lee, D.; Kang, J.J.; Joo, H.T.; Lee, J.H.; Lee, H.W.; Lee, S.H. The effects of different environmental factors on the biochemical composition of particulate organic matter in Gwangyang Bay, South Korea. Biogeosciences 2017, 14, 1903–1917. [Google Scholar] [CrossRef] [Green Version]
  15. Smith, W.O., Jr.; Ainley, D.G.; Arrigo, K.R.; Dinniman, M.S. The oceanography and ecology of the Ross Sea. Annu. Rev. Mar. Sci. 2014, 6, 469–487. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  16. Jones, R.M.; Smith, W.O., Jr. The influence of short-term events on the hydrographic and biological structure of the southwestern Ross Sea. J. Mar. Syst. 2017, 166, 184–195. [Google Scholar] [CrossRef] [Green Version]
  17. Peloquin, J.A.; Smith, W.O., Jr. Phytoplankton blooms in the Ross Sea, Antarctica: Interannual variability in magnitude, temporal patterns, and composition. J. Geophys. Res. Oceans 2007, 112, C08013. [Google Scholar] [CrossRef] [Green Version]
  18. Mangoni, O.; Saggiomo, V.; Bolinesi, F.; Margiotta, F.; Budillon, G.; Cotroneo, Y.; Misic, C.; Rivaro, P.; Saggiomo, M. Phytoplankton blooms during austral summer in the Ross Sea, Antarctica: Driving factors and trophic implications. PLoS ONE 2017, 12, e0176033. [Google Scholar] [CrossRef] [Green Version]
  19. Smith, W.O., Jr.; Kaufman, D.E. Climatological temporal and spatial distributions of nutrients and particulate matter in the Ross Sea. Prog. Oceanogr. 2018, 168, 182–195. [Google Scholar] [CrossRef]
  20. Xing, X.; Boss, E.; Chen, S.; Chai, F. Seasonal and Daily-Scale Photoacclimation Modulating the Phytoplankton Chlorophyll-Carbon Coupling Relationship in the Mid-Latitude Northwest Pacific. J. Geophys. Res. Oceans 2021, 126, e2021JC017717. [Google Scholar] [CrossRef]
  21. LaRue, M.A.; Ainley, D.G.; Swanson, M.; Dugger, K.M.; Lyver, P.O.; Barton, K.; Ballard, G. Climate Change Winners: Receding Ice Fields Facilitate Colony Expansion and Altered Dynamics in an Adélie Penguin Metapopulation. PLoS ONE 2013, 8, e60568. [Google Scholar] [CrossRef]
  22. Bracegirdle, T.J.; Stephenson, D.B. Higher precision estimates of regional polar warming by ensemble regression of climate model projections. Clim. Dyn. 2012, 39, 2805–2821. [Google Scholar] [CrossRef]
  23. Yamada, K.; Ishizaka, J. Estimation of interdecadal change of spring bloom timing, in the case of the Japan Sea. Geophys. Res. Lett. 2006, 33, L02608. [Google Scholar] [CrossRef]
  24. Germán, A.; Andreo, V.; Tauro, C.; Scavuzzo, C.M.; Ferral, A. A novel method based on time series satellite data analysis to detect algal blooms. Ecol. Inform. 2020, 59, 101131. [Google Scholar] [CrossRef]
  25. Kaufman, D.E.; Friedrichs, M.A.; Smith, W.O., Jr.; Queste, B.Y.; Heywood, K.J. Biogeochemical variability in the southern Ross Sea as observed by a glider deployment. Deep. Sea Res. Part I Oceanogr. Res. Pap. 2014, 92, 93–106. [Google Scholar] [CrossRef]
  26. Kaufman, D.E.; Friedrichs, M.A.; Smith, W.O., Jr.; Hofmann, E.E.; Dinniman, M.S.; Hemmings, J.C. Climate change impacts on southern Ross Sea phytoplankton composition, productivity, and export. J. Geophys. Res. Oceans 2017, 122, 2339–2359. [Google Scholar] [CrossRef]
  27. Friedland, K.D.; Mouw, C.B.; Asch, R.G.; Ferreira, A.S.A.; Henson, S.; Hyde, K.J.W.; Morse, R.E.; Thomas, A.C.; Brady, D.C. Phenology and time series trends of the dominant seasonal phytoplankton bloom across global scales. Glob. Ecol. Biogeogr. 2018, 27, 551–569. [Google Scholar] [CrossRef]
  28. Arrigo, K.R.; van Dijken, G.L. Annual changes in sea-ice, chlorophyll a, and primary production in the Ross Sea, Antarctica. Deep. Sea Res. Part II Top. Stud. Oceanogr. 2004, 51, 117–138. [Google Scholar] [CrossRef]
  29. Reddy, T.E.; Arrigo, K.R. Constraints on the extent of the Ross Sea phytoplankton bloom. J. Geophys. Res. Earth Surf. 2006, 111, C07005. [Google Scholar] [CrossRef] [Green Version]
  30. Park, J.; Kim, J.H.; Kim, H.C.; Hwang, J.; Jo, Y.H.; Lee, S.H. Environmental forcings on the remotely sensed phytoplankton bloom phenology in the central Ross Sea polynya. J. Geophys. Res. Oceans 2019, 124, 5400–5417. [Google Scholar] [CrossRef]
  31. Chen, S.; Smith, W.O., Jr.; Yu, X. Revisiting the Ocean Color Algorithms for Particulate Organic Carbon and Chlorophyll-a Concentrations in the Ross Sea. J. Geophys. Res. Oceans 2021, 126, e2021JC017749. [Google Scholar] [CrossRef]
  32. Szeto, M.; Werdell, P.J.; Moore, T.S.; Campbell, J.W. Are the world’s oceans optically different? J. Geophys. Res. Oceans 2011, 116, C00H04. [Google Scholar] [CrossRef] [Green Version]
  33. Mitchell, B. Predictive bio-optical relationships for polar oceans and marginal ice zones. J. Mar. Syst. 1992, 3, 91–105. [Google Scholar] [CrossRef]
  34. Matsuoka, A.; Huot, Y.; Shimada, K.; Saitoh, S.-I.; Babin, M. Bio-optical characteristics of the western Arctic Ocean: Implications for ocean color algorithms. Can. J. Remote Sens. 2007, 33, 503–518. [Google Scholar] [CrossRef]
  35. Matsuoka, A.; Hill, V.; Huot, Y.; Babin, M.; Bricaud, A. Seasonal variability in the light absorption properties of western Arctic waters: Parameterization of the individual components of absorption for ocean color applications. J. Geophys. Res. Earth Surf. 2011, 116, C02007. [Google Scholar] [CrossRef] [Green Version]
  36. O’Reilly, J.E.; Maritorena, S.; Siegel, D.A.; O’Brien, M.C.; Toole, D.; Mitchell, B.G.; Culver, M. Ocean color chlorophyll a algorithms for SeaWiFS, OC2, and OC4: Version 4. SeaWiFS Postlaunch Calibration Valid. Anal. Part 2000, 3, 9–23. [Google Scholar]
  37. Hu, C.; Lee, Z.; Franz, B. Chlorophyll a algorithms for oligotrophic oceans: A novel approach based on three-band reflectance difference. J. Geophys. Res. Oceans 2012, 117, C01011. [Google Scholar] [CrossRef] [Green Version]
  38. Le, C.; Zhou, X.; Hu, C.; Lee, Z.; Li, L.; Stramski, D. A color-index-based empirical algorithm for determining particulate organic carbon concentration in the ocean from satellite observations. J. Geophys. Res. Oceans 2018, 123, 7407–7419. [Google Scholar] [CrossRef]
  39. Smith, W.O., Jr.; Shields, A.R.; Peloquin, J.A.; Catalano, G.; Tozzi, S.; Dinniman, M.S.; Asper, V.A. Interannual variations in nutrients, net community production, and biogeochemical cycles in the Ross Sea. Deep. Sea Res. Part II Top. Stud. Oceanogr. 2006, 53, 815–833. [Google Scholar] [CrossRef]
  40. Park, J.; Kim, H.-C.; Bae, D.; Jo, Y.-H. Data Reconstruction for Remotely Sensed Chlorophyll-a Concentration in the Ross Sea Using Ensemble-Based Machine Learning. Remote Sens. 2020, 12, 1898. [Google Scholar] [CrossRef]
  41. Toseland, A.; Daines, S.J.; Clark, J.R.; Kirkham, A.; Strauss, J.; Uhlig, C.; Lenton, T.M.; Valentin, K.; Pearson, G.A.; Moulton, V.; et al. The impact of temperature on marine phytoplankton resource allocation and metabolism. Nat. Clim. Change 2013, 3, 979–984. [Google Scholar] [CrossRef]
  42. Liu, X.; Smith, W.O. Physiochemical controls on phytoplankton distributions in the Ross Sea, Antarctica. J. Mar. Syst. 2012, 94, 135–144. [Google Scholar] [CrossRef]
  43. Smith, W.O., Jr.; Jones, R.M. Vertical mixing, critical depths, and phytoplankton growth in the Ross Sea. ICES J. Mar. Sci. 2015, 72, 1952–1960. [Google Scholar] [CrossRef] [Green Version]
  44. Smith, W.O., Jr.; McGillicuddy, D.J., Jr.; Olson, E.B.; Kosnyrev, V.; Peacock, E.E.; Sosik, H.M. Mesoscale variability in intact and ghost colonies of Phaeocystis antarctica in the Ross Sea: Distribution and abundance. J. Mar. Syst. 2017, 166, 97–107. [Google Scholar] [CrossRef] [Green Version]
  45. Yuan, X. ENSO-related impacts on Antarctic sea ice: A synthesis of phenomenon and mechanisms. Antarct. Sci. 2004, 16, 415–425. [Google Scholar] [CrossRef]
  46. Stammerjohn, S.E.; Martinson, D.G.; Smith, R.C.; Yuan, X.; Rind, D. Trends in Antarctic annual sea ice retreat and advance and their relation to El Niño–Southern Oscillation and Southern Annular Mode variability. J. Geophys. Res. Earth Surf. 2008, 113, C03S90. [Google Scholar] [CrossRef] [Green Version]
Remotesensing 14 03263 g001 550
Figure 1. Comparison of the spatial distribution of the phytoplankton blooms on 12–18 February 2005, identified using the locally-tuned Chl algorithm (a,b) and the NASA standard Chl algorithm (c,d), respectively. Note that the green patches in (b,d) are the bloom masks corresponding to the Chl data in (a,c), and blue patches are observations without blooms.
Figure 1. Comparison of the spatial distribution of the phytoplankton blooms on 12–18 February 2005, identified using the locally-tuned Chl algorithm (a,b) and the NASA standard Chl algorithm (c,d), respectively. Note that the green patches in (b,d) are the bloom masks corresponding to the Chl data in (a,c), and blue patches are observations without blooms.
Remotesensing 14 03263 g001
Remotesensing 14 03263 g002 550
Figure 2. The spatial distribution of phytoplankton blooms in the Ross Sea in terms of bloom frequency between 2002 and 2021. Bloom frequency was quantified by dividing the total number of bloom observations by the total number of valid satellite observations, which was used to indicate the possibility of phytoplankton blooms to occur in a certain region. Phytoplankton blooms were defined with a Chl threshold of 2.0 mg m−3, based on the locally-tuned Chl algorithm. Note that the white gaps suggest few blooms ever occurred.
Figure 2. The spatial distribution of phytoplankton blooms in the Ross Sea in terms of bloom frequency between 2002 and 2021. Bloom frequency was quantified by dividing the total number of bloom observations by the total number of valid satellite observations, which was used to indicate the possibility of phytoplankton blooms to occur in a certain region. Phytoplankton blooms were defined with a Chl threshold of 2.0 mg m−3, based on the locally-tuned Chl algorithm. Note that the white gaps suggest few blooms ever occurred.
Remotesensing 14 03263 g002
Remotesensing 14 03263 g003 550
Figure 3. (a) Observation frequency of valid satellite observations in the Ross Sea between 2002 and 2020 from MODIS. (b) Same as Figure 2, but bloom frequency derived using the NASA standard Chl algorithm.
Figure 3. (a) Observation frequency of valid satellite observations in the Ross Sea between 2002 and 2020 from MODIS. (b) Same as Figure 2, but bloom frequency derived using the NASA standard Chl algorithm.
Remotesensing 14 03263 g003
Remotesensing 14 03263 g004 550
Figure 4. Interannual variations of the phytoplankton bloom areas (a), bloom strength (b), and Chl (c) between 2002 and 2021 in the Ross Sea. Note that the black and red curves are derived using the NASA standard algorithm and the locally tuned Rrs667-based algorithm, respectively, and the error bar associated refers to one standard deviation. The bloom strength was quantified for pixels with Chl ≥ 2 mg m−3.
Figure 4. Interannual variations of the phytoplankton bloom areas (a), bloom strength (b), and Chl (c) between 2002 and 2021 in the Ross Sea. Note that the black and red curves are derived using the NASA standard algorithm and the locally tuned Rrs667-based algorithm, respectively, and the error bar associated refers to one standard deviation. The bloom strength was quantified for pixels with Chl ≥ 2 mg m−3.
Remotesensing 14 03263 g004
Remotesensing 14 03263 g005 550
Figure 5. Interannual variations of SST (a), PAR (b), SSW (c), Wdir (d), and SIC (e) between 2002 and 2021 in the Ross Sea.
Figure 5. Interannual variations of SST (a), PAR (b), SSW (c), Wdir (d), and SIC (e) between 2002 and 2021 in the Ross Sea.
Remotesensing 14 03263 g005
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Chen, S.; Meng, Y. Phytoplankton Blooms Expanding Further Than Previously Thought in the Ross Sea: A Remote Sensing Perspective. Remote Sens. 2022, 14, 3263. https://doi.org/10.3390/rs14143263

AMA Style

Chen S, Meng Y. Phytoplankton Blooms Expanding Further Than Previously Thought in the Ross Sea: A Remote Sensing Perspective. Remote Sensing. 2022; 14(14):3263. https://doi.org/10.3390/rs14143263

Chicago/Turabian Style

Chen, Shuangling, and Yu Meng. 2022. "Phytoplankton Blooms Expanding Further Than Previously Thought in the Ross Sea: A Remote Sensing Perspective" Remote Sensing 14, no. 14: 3263. https://doi.org/10.3390/rs14143263

APA Style

Chen, S., & Meng, Y. (2022). Phytoplankton Blooms Expanding Further Than Previously Thought in the Ross Sea: A Remote Sensing Perspective. Remote Sensing, 14(14), 3263. https://doi.org/10.3390/rs14143263

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop