Spatio-temporal variation of microphytoplankton communities in Obhur Creek, the central Red Sea

In this study, three sites were selected for monthly sampling within the creek. Site 1 (reference site) was 2 km away from the creek mouth toward the open waters and is expected to be away from any source of anthropogenic impact. It has an average depth of about 200 m and shows typical characteristics of the Red Sea coastal waters. Site 2 was located in the middle zone of the creek and receives discharges from an aquaculture facility. Site 3 was located at the northeastern end of the creek, exposed to human disturbance, and was characterized by shallow waters and weak water exchange (Fig. 1).

Figure 1

Monthly phytoplankton and surface seawater samples were collected from January through December (2017). The sampling was carried out during daytime using a mechanized boat. Salinity and temperature were measured in situ using a water quality probe (Horiba U50). A Niskin sampler (Hydrobios – 5 l) was used to collect 10 l of seawater from a depth of 0.5 m in order to measure inorganic nutrients and phytoplankton biomass (chlorophyll a). To determine chlorophyll a, 3–5 l of seawater was filtered through a Whatman GF/F filter paper (0.7 μm, 47 mm) and kept at −80°C until further analysis. To estimate inorganic nutrient concentrations, 500 ml of seawater was filtered through a Whatman Nucleopore membrane filter (0.2 μm). The analysis of both chlorophyll a and inorganic nutrients (nitrate, nitrite, ammonia, phosphate and silicate) was carried out according to the protocols by Parsons et al. (1984) using a UV spectrophotometer (Shimadzu UV 1700).

A phytoplankton net (Hydrobios) with a mesh size of 20 μm was used for sampling. The net was fitted with a flowmeter to calculate the volume of filtered water (VWF) based on this equation: VWF=πr2×d VWF = \pi {r^2} \times d where “r” is the radius of the net and “d” is the haul distance, which is obtained as the difference between initial and final flowmeter readings. The net was towed horizontally for 6–10 min at a boat speed of ~1 knot. The collected samples were then immediately fixed with Lugol's iodine solution and a few drops of concentrated formaldehyde solution (Kürten et al. 2015). Prior to analysis, samples were screened through a 500 μm net to remove large particles of both biological and non-biological origin. A Sedgewick Rafter Counting Cell (1 ml/1 μl) was used to assess phytoplankton abundance under an inverted microscope (Leica DMI 3000B). The protocols provided by LeGresley & McDermott (2010) were followed for systematic analysis of phytoplankton abundance and a triplicate counting procedure was performed on each phytoplankton sample to increase the accuracy of the analysis. The counting method detailed in Devassy et al. (2019) was employed to determine the number of squares in Sedgewick-Rafter chamber. Phytoplankton species were taxonomically classified with the help of identification catalogues (Taylor 1976; Tomas 1997; Hallegraeff 2003; Gómez 2013) and then validated with the help of WoRMS (World Register of Marine Species; www.marinespecies.org) and named according to the latest taxonomical nomenclature.

Relationships between physicochemical variables and phytoplankton biomass and abundance were determined using Pearson's correlation coefficient r (SPSS V23). One-way analysis of variance (ANOVA) was performed to determine spatial and temporal variations of different parameters using SPSS V23. Species richness, Shannon–Wiener index (H’) and evenness (J’) as well as the Bray–Curtis Similarity Index were computed using PRIMER 6 (Clarke & Gorley 2006). Prior to analysis, the data were square root transformed due to the apparent deviation from the normal distribution.

Spatial variation in temperature distribution was less pronounced, while significant temporal variation (p < 0.01) in salinity was observed during the study. Maximum temperature (32.2°C) and salinity (40.18) were recorded at site 3 in September, while the minimum values (24.5°C and 38.85) were recorded in March at sites 1 and 3, respectively (Fig. 2). Salinity values at site 1 were almost similar throughout the year, ranging between 38.85 and 39.52 in March and August, respectively (Fig. 2). Site 3 differed slightly and showed higher values (average: 39.53 ± 0.40) compared to the two other sites.

Figure 2

Nitrate concentrations (NO₃⁻) ranged between a minimum of 0.03 μmol l⁻¹ at site 2 in June and a maximum of 4.50 μmol l⁻¹ at site 3 in August with an overall average of 0.73 ± 0.76 μmol l⁻¹. Although only slight variations in nitrate concentration were observed at sites 1 and 2 (mean values: 0.52 ± 0.60 and 0.34 ± 0.23 μmol l⁻¹, respectively), site 3 showed higher nitrate concentration (average: 1.32 ± 1.41 μmol l⁻¹; Fig. 3a). Higher nitrate values were observed at site 3 between August and November (Fig. 3a), with a maximum of 4.50 μmol l⁻¹ in August. Nitrite (NO₂⁻) values ranged from 0.01 to 0.22 μmol l⁻¹ at site 1 (August) and site 3 (September), respectively (Fig. 3b). On the other hand, ammonia (NH₄⁺) showed significant variations among the sites. Higher ammonia concentration was observed at site 3 (average: 1.48 ± 1.31 μmol l⁻¹) followed by sites 2 and 1 (mean values: 1.04 ± 0.99 μmol l⁻¹ and 0.36 ± 0.21 μmol l⁻¹, respectively). Similar to nitrate, ammonia values also showed an increasing trend toward the second half of the year in the study region (Fig. 3c). Phosphate (PO₄³⁻) ranged between 0.01 and 0.33 μmol l⁻¹ with an average of 0.08 ± 0.08 μmol l⁻¹ and no spatial variation was detected. Higher phosphate concentrations were observed between February and April with a maximum average value of 0.25 ± 0.07 μmol l⁻¹ in March (Fig. 3d). Silicate concentration varied significantly among the sites, with relatively higher values at site 3 (average: 2.38 ± 0.38 μmol l⁻¹) compared to the other sites (Fig. 3e).

Figure 3

Chlorophyll a concentration was generally low throughout the study region (average: 0.35 ± 0.32 mg m⁻³) with a pattern similar to the distribution of nutrients, except a few cases (Fig. 3f). Chlorophyll a concentrations were relatively similar at sites 1 and 2 (mean values: 0.39 ± 0.48 and 0.40 ± 0.27 mg m⁻³, respectively), though site 3 recorded the lowest mean value (0.26 ± 0.22 mg m⁻³). The highest chlorophyll concentration (1.85 mg m⁻³) was determined at site 1 in May, while the lowest value (0.015 mg m⁻³) was recorded at site 3 in May. Chlorophyll values ranged between 0.05 and 1.85 mg m⁻³ at site 1 in April and May, respectively, between 0.05 and 1.04 mg m⁻³ at site 2 in January and May, respectively, and between 0.015 and 0.85 mg m⁻³ at site 3 in May and December, respectively (Fig. 3f).

Diatoms were by far the most abundant group in the phytoplankton communities, accounting for 14.1 to 97% of the total phytoplankton throughout the year (mean: 75%). Dinoflagellates and cyanophytes, accounting for 20% (2.6–85.7%) and 5% (0.1–45.1%) of the total phytoplankton, ranked second and third in the abundance, respectively. Phytoplankton community composition was characterized by high diversity during the study period, with a total of 220 species (Supplementary material 1). Of the 220 phytoplankton species, 117 belonged to diatoms (76 Centrales and 41 Pennales), 99 to dinoflagellates and four species to cyanophytes. The maximum number of species (174) was found at site 2, which was followed by site 1 (170) and site 3 (128). Sites 1 and 2 were characterized by approximately similar diversity, with 84 diatoms and 81 dinoflagellates at the former (site 1) and 94 diatoms and 76 dinoflagellates at the latter (site 2) observed throughout the study period. Site 3 was less diverse than the two other sites, with 62 diatom and 64 dinoflagellate species (Supplementary material 1). Four Cyanophyta species were recorded at site 2 and only two species were found at sites 1 and 3, with the dominance of Trichodesmium sp., which accounted for 89.9% of the total cyanophytes. Centric diatoms were more dominant than pennates throughout the study period. Of the total phytoplankton species (220) observed, 21 diatoms and four dinoflagellate species are new records for the Red Sea. In addition, two diatoms (including Pseudo-nitzschia cf. delicatissima) and 14 dinoflagellates were listed on the IOC-UNESCO taxonomic reference list (Supplementary material 1) as potentially harmful species worldwide.

Rhizosolenia and Chaetoceros were the most diverse diatom genera in the current study (16 and 15 species, respectively). Other important diatom genera that contributed significantly to the diversity were: Pleurosigma (four species), Cerataulina, Guinardia, Navicula and Nitzchia (three species each; Supplementary material 1). Among dinoflagellates, the orders Gonyaulacales and Peridiniales were represented by the maximum number of genera (six genera), followed by Dinophysiales (three genera) and Gymnodiniales (two genera). The genus Tripos (synonym Ceratium) of dinoflagellates was observed with a maximum number of species (27 species), and was followed by Protoperidinium (18 species) and Dinophysis (nine species). At site 3, heterotrophic dinoflagellate species (33 species) dominated over the autotrophic ones and, for most of the study period, over diatoms. The most common dinoflagellate species occurring throughout the study period were: Dinophysis caudata, D. tripos, Protoperidinium conicum, P. divergens, P. steinii, Tripos furca, T. fusus, T. horridus, T. lineatus and T. teres (Supplementary material 1). On the other hand, the cyanophyte Trichodesmium sp. was recorded in relatively large numbers at sites 1 and 2 in June and July, and at site 3 in July.

The abundance of phytoplankton in the study area varied between 7.95×10³ and 3063.27 × 10³ individuals m⁻³, with an overall average of 238.59 × 10³ ± 540.38 × 10³ individuals m⁻³. The highest average abundance of total phytoplankton was recorded at sites 1 and 2, with 295.72 × 10³ ± 868.36 × 10³ and 231.44 × 10³ ± 306.98 × 10³ individuals m⁻³, respectively. On the other hand, the lowest average abundance was recorded at site 3 (139.81 × 10³ ± 254.43 × 10³ individuals m⁻³). Regarding the monthly variation, a sharp peak in abundance was observed in May (average: 1389.63 × 10³ ± 1543.12 × 10³ individuals m−3) due to the high density of Pseudo-nitzschia cf. delicatissima at sites 1 and 2 (2985.5 × 10³ and 969.04 × 10³ individuals m⁻³, accounting for 97.5% and 89.5% of the total phytoplankton at these sites in May, respectively). Moreover, small increases were detected in July, October and December with average density of 256.62 × 10³, 267.45 × 10³ and 356.26 × 10³ individuals m⁻³, respectively. The total monthly phytoplankton abundance at site 1 varied between 7.95 × 10³ individuals m⁻³ in September and 3063.27 × 10³ individuals m⁻³ in May (Fig. 4a). At site 2, the abundance fluctuated between 14.48 × 10³ and 1082.34 × 10³ individuals m⁻³ in January and May, respectively (Fig. 4c). The abundance at site 3 ranged between 23.26 × 10³ individuals m⁻³ in May and 929.34 × 10³ individuals m⁻³ in December (Fig. 4e).

Figure 4

With an average abundance of 280.20 × 10³ ± 861.17 individuals m⁻³ and a percentage contribution ranging from 21.05% in June to 98.36% in May, diatoms significantly dominated in the total phytoplankton abundance at site 1 for most of the study period. Dinoflagellates (average abundance: 24.27 × 10³ ± 29.52 individuals m⁻³), on the other hand, significantly dominated in the phytoplankton community at this site in August (64.91%) and September (52.50%). Cyanophytes (average abundance: 15.59 × 10³ ± 43.17 individuals m⁻³) dominated only in June and July, accounting for 60.15% and 46.55%, respectively. On the other hand, their contribution to the total abundance in the other months was almost negligible (Fig. 4b). Site 2 also showed a similar pattern of phytoplankton distribution as site 1, with diatoms (average abundance: 174.25 × 10³ ± 290.93 individuals m⁻³) being the most abundant group for most of the study period, except summer (July–September) when dinoflagellates (average abundance: 55.53 × 10³ ± 52.45 individuals m⁻³) dominated in the total phytoplankton abundance. Diatoms contributed 9.24% and 94.21% to the total abundance in July and May, respectively, while dinoflagellates contributed between 4.68% and 60.78% to the total abundance in May and September, respectively. Despite the presence of cyanophytes (average abundance: 17.02 × 10³ ± 47.8 individuals m⁻³) in summer (June–July), they were not the most abundant phytoplankton group (30.30 and 42.44%) in these particular months and showed negligible presence in the other ones (Fig. 4d). Site 3 differed from the two other sites in terms of the distribution of various phytoplankton groups and it was dominated by dinoflagellates (average abundance: 63.44 × 10³ ± 69.67 individuals m⁻³) for most of the study period, with a contribution ranging from 14% (June) to 92.08% (January). Diatoms (average abundance: 82.40 × 10³ ± 204.97 individuals m⁻³) dominated in May–June and November–December and contributed from about 7.92% (January) to 86% (June) to the total phytoplankton abundance. Cyanophytes (average abundance: 3.07 × 10³ ± 8.52 individuals m⁻³) dominated at this site (site 3) in July (56.25%) and were scarce in the other months (Fig. 4f).

Values of Pearson's coefficient of correlation (r) between individual parameters obtained in the study period showed a significant correlation for the total phytoplankton abundance with chlorophyll concentration (p < 0.01) as well as with silicate concentration (p < 0.05; Table 1). Physical parameters and nutrient salts showed a non-significant correlation with phytoplankton biomass and abundance. One-way analysis of variance (ANOVA) of salinity, nitrite, ammonia and silicate showed significant spatial variation, while temperature, salinity and phosphate showed significant temporal variation (Table 2). The dendrogram based on the Bray–Curtis similarity index of mean phytoplankton abundance showed a relatively high similarity between the sampling sites (> 60%). The similarity matrix of monthly phytoplankton abundance at each site also showed a roughly similar distribution pattern between the sites (Fig. 5). The dendrogram of site 1 (Fig. 5a) showed almost 78% similarity between the months except May, which was clearly separated from the others (35%). At site 2, the dendrogram indicated 63% similarity between the months (Fig. 5b). At site 3, on the other hand, all months showed 74% similarity except December (54%; Fig. 5c). Biodiversity indices clearly showed differences in the number of species. The lowest number of species was always recorded at site 3 (average: 28 ± 10), while sites 1 and 2 contained larger numbers (mean values: 44 ± 11 and 46 ± 17 at sites 1 and 2, respectively; Table 3). The Shannon–Wiener diversity index (H’) further confirmed differences in the diversity between the sites by providing relatively higher indices for site 1 (average: 3.75 ± 0.25) and 2 (average: 3.77 ± 0.35) compared to site 3 (average: 3.3 ± 0.32; Table 3). Relatively similar Pielou's evenness index (J’) revealed a uniform distribution of phytoplankton species at the study sites, with values ranging in a narrow range between 0.977 and 1.000 (Table 3).

Figure 5