First record of straight-needle pteropod Creseis acicula Rang, 1828 bloom in the Çanakkale Strait (NE Aegean Sea, Turkey)

Marine environmental parameters (sea surface temperature, salinity, dissolved oxygen and pH) of the sampling locations were determined using a CTD device. The parameters were compared with previous measurements conducted in the same region during the Marmara Sea Environmental Monitoring (MAREM) Project between 2007 and 2018 (Fig. 5; see MAREM reports). Pearson correlation analysis was applied to detect possible correlations between the variables (Table 3). The correlation analysis includes the dataset for the period between 2008 and 2018, excluding 2007 due to data availability at sampling site 6c, which is the focus of this study.

Figure 5

Nitrogen is one of the most important nutrients regulating the primary production of phytoplankton, which are grazed by zooplankton (Bristow et al. 2017; Sitta et al. 2018). In the study area, nitrate (NO₃⁻), i.e. the inorganic form of nitrogen, is found in higher concentrations compared to other nutrient sources such as nitrite and phosphate (Tuğrul et al. 2002). To investigate changes in nutrient concentrations, monthly mean nitrate values obtained using the MedBFM (Mediterranean Sea Biogeochemistry Analysis and Forecast) model system were examined for the period from April 2017 to December 2020 (Fig. 6) and compared with chlorophyll-a concentrations (Figs 7 and 8).

Figure 6

Figure 7

Figure 8

Chlorophyll-a (Chl-a) is the major photosynthetic pigment of phytoplankton living in aquatic environments, and is considered the most suitable proxy for phytoplankton biomass studies related to primary production (Huot et al. 2007), which is affected by nutrient supply, i.e. nitrate concentration in seawater (Cermeño et al. 2016). Therefore, it gives us an insight into biochemical variability in the vicinity of the sampling site 6c (Figs 7 and 8).

The data used in this study were derived from monthly and daily interpolated Chl-a concentrations in the Mediterranean Sea obtained from multi-satellite and Sentinel-3 OLCI observations distributed by the Global Ocean Satellite monitoring and marine ecosystem study group of the Italian National Research Council in Rome.

Both datasets (nitrate and Chl-a concentrations) can be accessed via the online platform of the E.U. Copernicus Marine Service (https://marine.copernicus.eu/).

Thirty-nine adult and mature individuals were measured for their morphological characteristics such as their total length and operculum width (Fig. 2, Fig. 3 and Table 2). Even though the juvenile forms were not included in the measurements, their occurrence with adult forms indicates a bloom event.

The length of the specimens varied between 7.05 and 13.03 mm and their operculum width varied between 0.36 and 0.73 mm. The regression analysis applied to test the correlation trend of the total length and the operculum width of the 39 measured specimens showed that they significantly correlate at r² = 0.67 with p < 0.05 (Fig. 5), indicating that 67% of the variation can be explained by these two explanatory variables.

Marine environmental parameters represent physicochemical conditions of the marine habitat in which organisms thrive. To investigate changes between previous and recent marine conditions, we have compared our measurements with data detailed in the MAREM reports from 2007 to 2018. We have restricted our comparison to the upper 0.5 m layer of surface water, where C. acicula blooms were observed in July 2020. We considered salinity, temperature, pH and dissolved oxygen values to analyze the environment of C. acicula blooms.

Sea surface temperatures (SSTs) measured at all sampling sites each year yield similar values, except for 2020 (Fig. 5a). SSTs at site 1c shows fluctuations between 22 and 28°C; site 2c between 23 and 28°C; site 3c shows the largest variations from 21 to 29°C; site 4c from 23.5 to 29°C; site 5c between 23 and 29.5°C, and site 6c between 22 and 28.5°C, with the highest temperature of 32°C measured in July 2020 (Fig. 5a).

Salinity values show fluctuations at each site (Fig. 5b). Site 1c shows variations between 21 and 24 PSU; site 2c between 21 and 28 PSU; site 3c and 4c between 19 and 28 PSU; site 5c from 20 to 24 PSU and site 6c from 4 to 28 PSU except for one measurement in 2010, which results in 22 PSU (Fig. 5b).

The pH measurements show large variations at each site each year (Fig. 5c). The pH values at sites 1c and 2c range from 7.5 to 8.4; at site 3c from 7.4 to 8.5; at site 4c from 7.1 to 8.4; at site 5c from 7.6 to 8.4, and site 6c shows the highest values ranging from 7.7 to 8.7, with the highest value of 8.7 measured in July 2020.

Dissolved oxygen (DO) values, except for those measured at site 4c in 2009 and site 5c in 2011, show a similar trend (Fig. 5d). DO at site 1c varies between 3.8 and 5.2 ml l⁻¹; at site 2c between 3.6 and 5.1 mmol mol⁻¹; at site 3c between 3.7 and 5.1 mmol mol⁻¹; at site 5c between 3.5 and 5.1 mmol mol⁻¹ and at site 6c between 4.2 and 5.3 mmol/mol, with the highest average value of 4.6 mmol mol⁻¹ (Fig. 5d).

Pearson's correlation results indicate no significant correlation at p < 0.05 between the physicochemical parameters (Table 3), suggesting an auxiliary variable inducing changes in the marine ecosystem of the study area.

The study area is characterized by fluctuating nutrient concentrations each season depending on the nutrient supply from the surrounding rivers and/or the two-layer water exchange system in the Çanakkale Strait (Fig. 6; Türkoğlu et al. 2006). From April 2017 to December 2018, it was represented by nitrate concentrations ranging from ~10 to 30 mmol m⁻³, with the highest values corresponding to spring and autumn-winter seasons. In April 2017, a concentration of ~23 mmol/m³ was recorded, steadily decreasing through August 2017 (Fig. 6a). The highest values were recorded between September and November 2017, showing a sharp decrease in January 2017, and from January 2017 until November 2018, showing a slight increase in December 2018 (Fig. 6a). No marked increase in nitrate concentrations was observed in spring of 2018.

From January 2019, on the other hand, these values dropped to the levels between ~0.6 and 1.6 mmol m⁻³ (Fig. 6b). Between February and March 2019, high peaks in nutrient concentrations of 1.5 to 1.6 mmol m⁻³ were recorded, corresponding to the highest values between January 2019 and December 2020. Nitrate concentration values steadily decreased between April 2019 and August 2019, with the lowest values < 0.8 mmol m⁻³. Concentration values showed an increase from September 2019 through April 2020, with fluctuating but high values between January 2020 and April 2020, which corresponds to the second peak interval between January 2019 and December 2020 (Fig. 6b). After April 2020, nitrate concentrations decreased again until September 2020 to values between 0.8 and 0.9 mmol m⁻³. July 2020, i.e. the sampling period of this study, falls within this range of decreasing nutrient concentrations with values slightly above 0.9 mmol m⁻³. After September 2020, they were increasing again until December 2020.

Chl-a concentration maps prepared for July from 2017 to 2020 demonstrate a parallel outcome to nitrate concentrations (Figs 6a and 6b), with higher values in July 2017 and July 2018, and lower values in July 2019 and July 2020 (Fig. 7). The years 2017 and 2018 were marked with higher Chl-a concentrations, with values in the range of 0.3–0.4 mg m⁻³, while values in July 2019 and July 2020 were around 0.1 mg m⁻³.

Interestingly, on the sampling day, the Sea of Marmara was characterized by a high nitrate concentration of about 0.3 mg m⁻³, which was not the case for the vicinity of the sampling area, where even lower values, i.e. below 0.1 mg m⁻³, were recorded (Fig. 8; 20 July 2020).

Morphological characteristics of pteropods depend on environmental parameters, such as dwarfing in size may be advantageous to increase the survival under stressed conditions (e.g. global warming, increased ocean acidity; Garilli et al. 2015; Roger et al. 2002).

The only available study involving measurements of total length and operculum width in specimens of C. acicula was conducted off the Australian coast, and it indicates a correlation between these two morphological features at r² = 0.85 (Roger et al. 2002). In our study, on the other hand, specimens correlate at r² = 0.67 (n = 39; p < 0.05). The total length of specimens reported in the study cited above ranged from 5 to 25 mm, and the operculum width ranged from 0.3 to 1.4 mm (Roger et al. 2002).

In comparison, Australian coasts are characterized by the following parameters: SSTs around 28°C, sea surface salinity around ~35 PSU, pH around 8.07, and DO around 4.5 ml l⁻¹ (CSIRO 2018; Locarnini et al. 2019; Zweng et al. 2019). On the other hand, site 6c in the Çanakkale Strait, where C. acicula was found in July 2020, was characterized by the following values: SST – 32°C, salinity – ~28 PSU, pH – 8.7 and DO – 5.05 ml l⁻¹. Even though the correlation details for total length and operculum width were not provided, the SSTs during the outbreak in Daya Bay varied between 29.8 and 32.8°C and salinity was > 30 PSU (Dai et al. 2020), which is higher than salinity values measured in July 2020 in the Çanakkale Strait. Similarly, during the C. acicula swarm in the southern Indian waters, the salinity was above 31 PSU (Pillai & Rodrigo 1979). Although it is difficult to infer due to scarcity of comparable data, it may be possible to consider salinity as a potential controlling factor for the growth pattern of C. acicula. Nonetheless, because C. acicula is one of the most common species in the subtropics and tropics (Bé & Gilmer 1977), a further study is needed to clarify the response of the growth pattern in C. acicula to physicochemical parameters of the surrounding waters.

The bloom, on the other hand, may have occurred as a result of a combination of several parameters. Site 6c, despite its proximity to warmer Aegean waters, shows lower SSTs compared to other sites along the Çanakkale Strait due to freshwater inflow from rivers (Türkoğlu et al. 2006). However, 20 July 2020 was characterized by a 9°C increase in ambient SST, from 23°C (in 2018) to 32°C (Fig. 5a). For comparison, high SST was also measured in the 2013 summer season (Fig. 5a). SST was ~28°C, but salinity was 24.5 PSU, and pH was 7.83 (Figs 5a, 5b and 5c, respectively). Even though DO was relatively high (5.33 ml l⁻¹; Fig. 5d), C. acicula was not found during sampling when physicochemical parameters were measured (Artüz et al. 2014), suggesting that DO may not be a factor controlling C. acicula blooms.

Since C. acicula was previously reported from the Southeastern Aegean waters and the Izmir coast, one can consider the possibility of a latitudinal migration of C. acicula associated with the improved surrounding seawater conditions sufficient to thrive. This theory would explain the monospecific bloom of this species. Other pelagic mollusk species, such as Limacina helicina in cold waters, were reported to migrate poleward as the temperature of surface ocean water increases (Comeau et al. 2012). Similarly, since C. acicula prefers warm and saline environments, increased SST due to global warming and increased salinity in the Çanakkale Strait along with increased pH levels may have triggered the migration of C. acicula from the Southern Aegean Sea to the northern coasts (Çanakkale Strait), which provide a suitable ecosystem for a bloom.

One important factor, in addition to physicochemical parameters, is the nutrient supply for a marine organism to thrive. The unique bloom of C. acicula that occurred in mid-July 2020 in the South China Sea was associated with the heavy rainfall season (Dai et al. 2020). Given its adaptability, C. acicula may have tolerated large salinity fluctuations recorded during the outbreak season due to heavy rainfall, which was considered to have provided an increased food supply and contributed to elevated Chl-a (Dai et al. 2020).

In comparison, nutrient concentrations in the study area may show seasonal fluctuations up to 10 –15 times (Fig. 6; Türkoğlu 2010), which is strongly affected by the two-layer water exchange system formed by warm, highly saline Mediterranean deep water and cooler, less saline surface water originating from the Black Sea as explained in Section 2. Nutrients in the surface water of the Black Sea to a depth of 20–40 m are mostly taken up within the Sea of Marmara, before they reach the Çanakkale Strait (Türkoğlu 2010), and the rest are dispersed in the Northeastern Aegean Sea, joining the surface currents (Figs 1 and 8).

It has been suggested that nutrient enrichment observed in different periods other than seasonal variations may be related to vertical mixing between the surface waters and nutrient-rich, oxygenated, saline waters of the Mediterranean Sea as they enter the Çanakkale Strait (Tuğrul et al. 2002; Türkoğlu 2010). However, in 2017 and 2018, monthly nitrate concentrations were > 10 mmol m⁻³ (Fig. 6a), whereas in 2019 and 2020, these values fluctuated between 0.6 and 1.6 mmol m⁻³ and were slightly above 0.9 mmol m⁻³ in July 2020 (Fig. 6b). The high Chl-a concentration in the Sea of Marmara, observed on the day of sampling (20 July 2020), did not continue through the Çanakkale Strait (Fig. 8), which may indicate nutrient uptake by phytoplankton and/or other marine organisms. This finding contradicts the food availability theory proposed for the outbreak recorded in Daya Bay, indicating a different controlling factor.

Researchers suggest that despite increased anthropogenic CO₂ concentrations, the Northern Aegean waters are not expected to be affected by carbonate ion depletion as they are supersaturated with both calcite and aragonite (Krasakopoulou et al. 2017). Interestingly, elevated pH and DO levels measured during the July 2020 sampling (Fig. 5c) indicate reduced acidity contradicting the statements regarding declining pH levels, elevated temperatures and pCO₂ on a global scale (Gruber 2011). Correspondingly, the simultaneous increase in these parameters (Figs 5c and 5d) may be related to reduced CO₂ emissions during the COVID-19 lockdowns between April and July 2020, reflecting reduced acidity in response to reduced greenhouse gas emissions, which most likely provided a suitable environment for these aragonite-shelled organisms to thrive.

The reduced CO₂ values in the first half of 2020 corresponding to COVID-19 lockdowns have been recently reported (Le Quéré et al. 2020; Liu 2020), and the Ministry of Environment and Urbanization of Turkey has reported a 17.4% daily reduction in greenhouse gas emissions at the country level during the COVID-19 restrictions (Gündoğmuş 2020). The daily reduction in greenhouse gas emissions in Turkey was estimated at 0.8% (9510 tons of CO₂) on 21 January 2020, and reached 17.4% (210 429 tons of CO₂) on 30 April 2020. In coastal areas, where the maritime transport sector operates, shipping emissions have been reported to contribute to GHG emissions, accounting for part of anthropogenic emissions (Winnes et al. 2015). Interestingly, in the region of the Sea of Marmara, 68% of the shipping emissions in the Turkish Straits are emitted in the Çanakkale Strait (Deniz & Durmuşoğlu 2008; Durmuşoğlu 2013).

According to the statistics obtained from the Directorate General of Maritime Commerce, the Ministry of Transport and Infrastructure of Turkey (https://atlantis.udhb.gov.tr), a total of 3692 and 3476 vessels respectively passed through the Çanakkale and Istanbul straits between April and July of 2017, 2018 and 2019 (Table 1). In 2020, a total of 3134 vessels from the Istanbul Strait and 3384 vessels from the Çanakkale Strait were recorded between April and July (Table 1), suggesting a decrease by ~10%. Consequently, the reduced domestic and international marine traffic in the Çanakkale Strait may have contributed to the reduction of greenhouse gas emissions along with the reduction of CO₂ release to surface waters, optimizing pH and dissolved oxygen levels in the sea surface waters for the bloom to occur.