Spatio-Temporal Dynamics of Phytoplankton Community Structure in Response to Environmental Drivers in Xiaohai Lagoon, Hainan Island, China

Abstract

The Xiaohai Lagoon is a vital coastal ecosystem that has faced decades of significant natural and anthropogenic pressures. This study investigated the spatio-temporal dynamics its phytoplankton communities through quarterly sampling from 2024 to 2025. Significant spatial and seasonal variations (p < 0.05) in physicochemical parameters were observed. The concentrations of various physicochemical parameters were highest at the lagoon mouth and decreased inwards. In contrast, sites inside the lagoon experienced elevated nutrient and organic matter indicators. Seasonally, the highest temperatures were recorded in Summer. However, Autumn recorded the highest NH₃-N and NO₂-N levels, while Winter recorded the highest NO₃-N levels. The findings generally suggest minimal pollution, as key physicochemical parameters, met the China water quality standard for environmental protection (GB 3838–2002). Overall, 109 phytoplankton species belonging to 38 genera and 5 phyla, including Cyanophyta, Bacillariophyta, Chlorophyta, Cryptophyta, and Dinophyta, were identified. The phytoplankton average density was 1.65 × 10³ Ind L⁻¹ with insignificant differences both spatially and seasonally (p > 0.05). One-way ANOSIM indicated significant seasonal dissimilarity in phytoplankton community composition (R = 0.828, p < 0.001), with SIMPER results revealing that Ceratocorys sp., Chaetoceros sp., Coscinodiscus subtilis, Oscillatoria princes, and Thalassionema nitzschioides contributed to the seasonal difference. CCA indicated phytoplankton composition and abundance were influenced by COD, TN, TDS, salinity, oxidation-reduction potential, EC, water temperature, NH₃-N, and NO₃-N. This study highlights the critical need for effective management strategies to protect and preserve the ecological integrity of Xiaohai Lagoon.

1. Introduction

Aquatic ecosystems are among Earth’s most biologically productive and ecologically significant systems, underpinning global biodiversity, climate regulation, and human livelihoods [1]. They create exceptionally productive habitats for a wide range of aquatic organisms, including phytoplankton, which form the foundation of aquatic food webs. Phytoplankton play an important role in primary productivity, energy flow, and maintaining the balance of aquatic ecosystems [2]. Additionally, they play a crucial role in the carbon cycle, contributing to geochemical and biological processes across continents. Phytoplankton exhibit particular sensitivity to changes in aquatic environments’ biotic, physical, and chemical (e.g., nutrient content) characteristics [3]. Moreover, their rapid response to environmental fluctuations makes them sensitive bio-indicators of water quality. For instance, the dominance by toxin-producing cyanobacteria or dinoflagellates can signal deteriorating water quality, cascading effects on food webs, and human health [4].

The phytoplankton communities’ structure, spatial patterns, and biodiversity are regulated by environmental factors, such as physical conditions, chemical properties, nutrient dynamics, and biological interactions [5]. Key drivers, including salinity, thermal gradients, light availability, hydrological turbulence, nutrient concentrations, and conductivity, play pivotal roles in shaping phytoplankton assemblages and their productivity in aquatic habitats [6]. Notably, nitrogen (N) and phosphorus (P) availability often act as primary constraints on phytoplankton biomass accumulation in marine and freshwater ecosystems. Variations in nutrient supply and the relative balance of N to P (N/P ratio) further modulate species-specific growth rates and population dynamics within these communities [3]. Factors like water turbidity, sediment load, particulate matter, and thermal stratification significantly influence phytoplankton community organization [7]. Environmental shifts can thus trigger reorganization of phytoplankton taxonomic profiles and functional diversity. These biological changes, in turn, may affect the ecosystem, altering biogeochemical cycles and habitat conditions.

Phytoplankton communities are increasingly vulnerable to anthropogenic pollution, which degrades aquatic ecosystems and disrupts their ecological balance [8]. Nutrients discharged from sewerage and agricultural runoff, coupled with rising temperatures and altered hydrology, disrupt phytoplankton community structure, favoring harmful algal blooms while suppressing beneficial species. Such disruptions destabilize ecosystems, reduce oxygen levels (hypoxia), and harm aquatic organisms. Coastal lagoons, in particular, face compounded threats due to their proximity to human settlements, making them hotspots for pollution, land-use changes, and hydrological alterations. These threats affect water temperature and nutrient availability, which regulate phytoplankton growth [9]. Therefore, it is essential to monitor the variation in phytoplankton community composition, diversity, and their influencing factors in anthropogenically affected areas. The Xiaohai Lagoon, a subtropical coastal lagoon found on the eastern bank of Hainan Island in the eastern part of Wanning City, in southern China, exemplifies these challenges. It is a biologically rich ecosystem, ecologically and socio-economically vital [10].

Nevertheless, the lagoon faces mounting anthropogenic pressures from urbanization, aquaculture expansion, and nutrient runoff from aquaculture and agricultural fields [11]. For decades, the lagoon has undergone dramatic morphological and hydrodynamic changes driven by both natural and anthropogenic activities, particularly from the large human settlement surrounding it. In recent years, with the discharge of aquaculture tailwater, agricultural runoff, and domestic sewage, the water quality of Xiaohai has deteriorated to aCategory 4 water body with an extremely poor status [12,13]. Luo investigated the spatial and temporal distribution of surface sediments in the Xiaohai lagoon [14]. Several studies have been conducted on Xiaohai Lagoon’s hydrodynamics, sediment transportation, morphological changes, and water quality [12,13,15,16]. For instance, Xingjian et al. [13] reported that the Xiaohai lagoon sediments were dominated by clayey silt, sandy silt, and silty sand. However, only a few studies have examined the dynamics of phytoplankton, which are important water quality indicators. Accordingly, this study was conducted to investigate the spatial and seasonal dynamics of phytoplankton community structure by characterizing the variations in water quality parameters, phytoplankton community composition, density, and diversity. To fully characterize the phytoplankton community distribution and environmental gradients, eight sampling sites were strategically selected encompassing key hydrological zones: the lagoon mouth, to monitor the seawater exchange and tidal influence; areas adjacent to primary riverine inputs, to assess the impact of terrestrial and potential anthropogenic runoff; and the middle of the lagoon basin, to evaluate pollutant accumulation and the intertidal areas which are susceptible to direct coastal and watershed influences.

2. Materials and Methods

2.1. Study Area

This study was conducted in the Xiaohai Lagoon, situated on the eastern bank of Hainan Island in the eastern part of Wanning City, and is connected to the South China Sea via a narrow inlet about 150 m [17]. It is the largest lagoon on Hainan Island, covering approximately 43.78 km², 7.5 km wide from east to west, and about 10 km long from north to south. The average water depth is 1.5 m, with a maximum depth of 4 m. The lagoon is also the confluence of rivers in the surrounding area; the main rivers that flow into it include the Taiyang, Longtou, Longwei, Dongshan, Longshou, Xigou, Beipo, and Baishi Rivers (Compiling Committee of Records of China Bays, 1999). The largest river is the Taiyang River (78.7 km), with an annual average runoff of 1.4 × 10⁸ m³ [18]. The watershed of the Taiyang River spans 593 km² and is primarily composed of agricultural areas. The Longtou River is 33.2 km long, and its watershed spans an area of 136 square km. The Longwei and Dongshan Rivers are 38.2 and 26.6 km long, with watersheds of 158 and 97 km², respectively. The Lagoon is in a tropical monsoon climate zone, with an average annual precipitation of 2159 mm. The average temperatures are approximately 28 °C in summer and 18 °C in winter. Xiaohai Lagoon has an irregular diurnal tide with a tidal range of 0.71 m at the entrance, which decreases inwards. The average salinity ranges from 4–5 in the lagoon to 8.1–31 at the entrance [13]. Considering the impact of human activities and ensuring thorough coverage across different parts of the lagoon for better sampling, eight sampling sites (Y1–Y8) were chosen to effectively represent the phytoplankton distribution in the lagoon, as shown in Figure 1.

2.2. Sample Collection and Analysis

The sampling was conducted from May 2024 to February 2025, covering the four seasons. The surface water samples were collected at a depth of 0.5 m below the water surface, using a 5 L water sampler. All samples were immediately transported to the laboratory and processed in accordance with the Chinese National Standard (HJ 493-2009) for preservation and analysis, with detailed methods outlined in

Table 1. Preservation methods and holding times for water quality parameters.

Parameter	Preservative	Storage Condition	Max Holding Time
COD	Concentrated H2SO4 to pH ≤ 2	Refrigeration (1–5 °C)	5 days
TP	Concentrated H2SO4 to pH ≤ 2	Refrigeration (1–5 °C)	24 h
TN	Concentrated H2SO4 to pH ≤ 2	Refrigeration (1–5 °C)	7 days
NH3-N	Concentrated H2SO4 to pH ≤ 2	Refrigeration (1–5 °C)	24 h
NO3-N	Hydrochloric Acid (HCl) to pH ≤ 2	Refrigeration (1–5 °C)	24 h
NO2-N, BOD5	None	Refrigeration (1–5 °C)	24 h
Chl-a	None	Refrigeration, dark (1–5 °C)	24 h

[19]. Water temperature (WT), pH, salinity (SAL), conductivity (EC), dissolved oxygen (DO), oxidation-reduction potential (ORP), total dissolved solids (TDS), and hydrostatic pressure (WP) were measured in situ using a YSI ProPlus (Professional Plus) multiparameter water quality meter (YSI Inc./Xylem Analytics, Yellow Springs, OH, USA). Probe accuracies were as follows: pH ± 0.2 units, temperature ± 0.01 °C, conductivity ± 0.5% reading, DO ± 0.1 mg L⁻¹, ORP ± 20 mV; salinity and TDS were calculated from conductivity, and pressure readings were convertible to mmHg. Transparency (SD) was determined with a Secchi disk, and water depth (WD) was measured with a depth sounder. The water samples collected were preserved using concentrated H₂SO₄ obtained from Xilong Chemical Co., Ltd., Shantou, Chinalabelled, and refrigerated or frozen, and immediately transported to the laboratory for analysis. pH, salinity (SAL, ppt), dissolved oxygen (DO, mg/L), conductivity (EC, µS/cm), temperature (WT, °C), pressure (WP, Hg), redox potential (ORP, mV), and total dissolved particulate matter (TDS, mg/L) were determined by water quality multi-parameter synthetic instrument, and transparency (SD, m) was determined by Selflat disk, and water depth (WD, m) was measured by depth meter. The determination of anodemic nutrients (NO₃-N, NO₂-N, NH₃-N, TN, TP), chemical oxygen demand (COD) and biochemical oxygen demand (BOD) was determined with reference to the Technical Specifications for Marine Monitoring, Part 4, Seawater Analysis, and the determination of chlorophyll a (Chl-a) was determined with reference to the Technical Specifications for Marine Monitoring, Part 7, Ecological Investigation and Biological Monitoring of Offshore Pollution. For the phytoplankton community structure, triplicate samples were obtained by filtering 20 L of sub-surface water through a 20 μm mesh plankton net. Immediately, the samples collected were fixed using neutral Lugol’s solution prepared according to the standard formulation: 10 g of potassium iodide and 5 g of iodine crystals dissolved in 100 mL of distilled water. The reagents, potassium iodide (Product No. P816709-100g, Macklin, Shanghai, China) and iodine (Product No. I812016-100g, Macklin, Shanghai), were purchased from Shanghai Macklin Biochemical Technology Co., Ltd. After fixing, the samples were concentrated in 30 mL specimen bottles after 48 h of sedimentation, according to Zhu et al. [20]. The concentrated samples (30 mL) were used to identify and quantify phytoplankton species. Approximately 0.1 mL of the samples was observed under a light microscope (×400) to identify and quantify the phytoplankton species. Plankton individuals were identified to the lowest taxonomic level using the standard identification keys [21,22,23]. To verify the taxonomic nomenclature of the phytoplankton species, all identified species were verified and updated against the World Register of Marine Species (WoRMS) available at https://www.marinespecies.org/index.php (accessed on 17 November 2025) to reflect the currently accepted names and synonyms. Finally, the phytoplankton abundance was extrapolated to individuals per litre (Ind L⁻¹)

2.3. Determination of Plankton Diversity and Dominance

The Shannon-Weiner’s diversity index, H, Pielou’s evenness index, J, and Margalef’s richness index, D, were used to evaluate the ecological characteristics of plankton [24]. Menhinick’s diversity index was calculated for phytoplankton [25].

$$\mathrm{H}=-{\sum }_{\mathrm{i}=1}^{\mathrm{s}}\left({\mathrm{P}}_{\mathrm{i}}\mathrm{l}\mathrm{n}{\mathrm{P}}_{\mathrm{i}}\right)$$

(1)

$$\mathrm{J}=\left(\mathrm{H}/{\mathrm{H}}_{\mathrm{m}\mathrm{a}\mathrm{x}}\right)$$

(2)

$$\mathrm{D}=\left(\mathrm{S}-1\right)/\mathrm{l}\mathrm{n}\mathrm{N}$$

(3)

where N is the number of total organisms in the samples, Pi is the proportion of individuals belonging to species i, Hmax is the maximum species diversity (Hmax = log₂S), and S is the total number of species in the samples.

In addition, McNaughton’s dominance index (Y) was used to analyze the dominant phytoplankton and zooplankton species [26].

$$\mathrm{Y}=\left({\mathrm{N}}_{\mathrm{i}}/\mathrm{N}\right)\times {\mathrm{f}}_{\mathrm{i}}$$

(4)

where Ni is the number of particular organism species i in the samples, N is the number of total organisms in the samples, and fi is the occurrence frequency of certain organisms in the samples. Y > 0.02 indicates the dominant species.

2.4. Statistical Analysis

Spatial distribution maps for physicochemical water quality parameters, phytoplankton density, and diversity were interpolated using ArcGIS 10.8 (ESRI, Redlands, CA, USA). At the same time, the spatial and seasonal variability in physicochemical water quality parameters were analyzed using one-way analysis of variance (ANOVA). Prior to ANOVA, data were checked for normality (Shapiro–Wilk test) and homogeneity of variances (Levene’s test). For variables that violated these assumptions, the non-parametric Kruskal–Wallis test was used as a robust alternative. Consequently, the Kruskal–Wallis test was applied to analyze variations in phytoplankton density and diversity across spatial and seasonal scales, as these data were often non-normal. Notably, before analysis, count data on phytoplankton abundance were log(x + 1) transformed, while the rest of the response variables were log-transformed to meet the normality assumptions. To test for significant differences in phytoplankton community composition between seasons, an analysis of similarities (ANOSIM) was performed. The dissimilarity among species was assessed using the SIMPER (similarity percentages) technique to determine the relative abundance of each contributing species. The seasonal distribution of phytoplankton species was visualized using principal component analysis (PCA). Detrended correspondence analysis (DCA) was first employed to determine whether linear or unimodal ordination methods were more suitable for the species data. The analysis of phytoplankton species using DCA revealed a maximum axis length of 5.24, greater than 3.0. Therefore, the unimodal ordination method, canonical correspondence analysis (CCA), was used to investigate the influence of environmental variables on phytoplankton species. The analyses were conducted using SPSS 28 and the vegan package in R 4.2.3 (R Core Team, 2019; https://www.R-project.org/).

3. Results and Discussion

3.1. Environmental Variables

The physicochemical and nutrient parameters of the sampling stations are summarized in Table S1 of Supplementary Information (SI). Generally, the average values were WT (25.59 °C), WP (759.04 mm Hg), DO (6.85), EC (31,777 µS cm−¹), TDS (20,333 mg L⁻¹), salinity (19.79 ppt), pH (8.22), ORP (94.76 mV), SD (0.93 m), WD (3.27 m), COD (1.30 mg L⁻¹), BOD (1.88 mg L⁻¹), NH₃-N (0.02 mg L⁻¹), NO₃-N (0.04 mg L⁻¹), NO₂-N (0.01 mg L⁻¹), TP (0.07 mg L⁻¹), TN (0.22 mg L⁻¹), and Chl-a (0.0048 mg L⁻¹). The TDS, salinity, SD, WD, BOD, TP, and Chl-a varied significantly across the sampling stations (p < 0.05, Table S1). Notably, the pH did not differ spatially and was within the MEP range of 6.5–8.5 for surface and natural waters to support aquatic life [27]. However, this study’s pH values were slightly lower than those reported in the Hangzhou Lagoon [28]. The DO concentration indicates the suitability of the water bodies to support the life of aquatic organisms. Notably, the average DO level in this study was 6.85, which is within the MEP limit (≥5.0 mg/L). Additionally, the NH₃-N, TP, and TN concentrations (Table S1) were within the limits of 0.2 mg L⁻¹ and ≤1 mg L⁻¹, respectively, set by the China Ministry of Environmental Protection (MEP) (GB 3838–2002) [29]. Likewise, the COD concentration in this study was within the MEP standards (20.0 mg L⁻¹). Additionally, Chl-a average concentration (0.0048 μg L⁻¹) was low compared to the MEP standards (<2.5 μg L⁻¹) but comparable to other studies [27,30]. Generally, these results reflected no severe pollution since the most critical parameters, such as pH, DO, COD, TN, and NH₃-N, met the China water quality standard for environmental protection (GB 3838–2002).

As demonstrated in the spatial distribution maps (Figure 2 and Table S1), the values of TDS, salinity, SD, and WD were highest at the lagoon mouth (Y1), followed by Y2 (near the lagoon mouth), while decreasing inside the lagoon. High TDS and salinity at the lagoon mouth may be associated with seawater intrusion, as the mouth serves as a direct connection to the sea, which is highly saline [31]. This could also be associated with tidal advection, as high tides in the sea constantly push seawater into the lagoon through the mouth [32]. Consequently, the water at the lagoon’s mouth is mainly undiluted or minimally diluted seawater, justifying this finding. The dominance of the seawater at the lagoon’s mouth could also explain the high water density observed at this sampling site, as the highly saline water entering the lagoon through the mouth is usually denser than the brackish water inside the lagoon [33,34]. Though not statistically significant, the lagoon mouth (Y1) also recorded the highest value of EC, which decreased inside the lagoon. Conductivity measures the ability to conduct electricity, usually dependent on the ion concentration (salts) [35]. Thus, high electrical conductivity could be directly linked to high salinity, and TDS observed at the lagoon’s mouth, as discussed above [36]. Interestingly, station Y6 (inside the lagoon) experienced elevated COD, BOD, TP, and Chl-a levels. In contrast, station Y1 (lagoon mouth) recorded the lowest values of these parameters (Table S1, Figure 2). This observation suggests nutrient enrichment, which may be attributed to anthropogenic inputs of nutrients primarily from agriculture, mariculture, or industrial waste [7], the primary activities in the Xiaohai lagoon [37]. Furthermore, the watersheds of the major rivers that feed the lagoon serve as agricultural areas, especially the Taiyang River watershed. Thus, the agricultural wastes packed with nutrients could find their way into the lagoon through runoff, elevating the nutrient levels. Since the lagoon is situated in the rapidly expanding Wanning City, a major coastal city in Hainan Province, on the eastern portion of Hainan Island, the sustained development of the Hainan Free Trade Port may also be the cause of the elevated nutrient concentration. This development has accelerated urbanization and agricultural intensification, significantly straining the environment [37,38,39]. The high Chl-a level observed at site Y6 is a direct result of the high nutrient levels, which could fuel algal growth. In contrast, high BOD levels could be attributed to high decomposition of organic matter at this site, as evidenced by the anoxic conditions observed. This site’s proximity to densely populated areas and croplands could expose it to severe pollution from sewage and agricultural runoff.

All physicochemical parameters showed notable seasonal variations except for pH, SD, WD, and Chl-a (Figure 3). High temperature values were recorded in May (29.41 °C) and August (28.99 °C), while low values were recorded in November (21.68 °C) and February (22.29 °C), driven by seasonal thermal dynamics, with high solar radiation in May and August (Summer). Increased solar radiation heats the water, leading to high temperatures during these months, and vice versa in February (Winter), when reduced solar radiation and increased precipitation prevail. August and November recorded the highest and lowest conductivity, TDS, and salinity levels, respectively (Figure 3), reflecting the lagoon’s interacting climatic, hydrological, and physical processes. High conductivity, TDS, and salinity recorded in August could be attributed to the intense evaporation that concentrates salts [40]. This season is also characterized by reduced river discharge into the lagoon, resulting in minimal dilution and higher salt concentrations, which increase conductivity. Furthermore, higher solar radiation during the summer (August) contributed to the high salinity, while rainfall-induced freshwater effect was probably responsible for the low salinity during the fall (November). Similar findings have been reported elsewhere [7]. On the other hand, November marks the onset of the wet season in Hainan, which increases the freshwater influx into the lagoon, resulting in dilution and a decrease in salt concentration. Therefore, the relatively high average conductivity recorded during summer (August) could be attributable to the increased salinity, suggesting a high level of dissolved salts. At the same time, November recorded the highest average values of NH₃-N (0.061 mg L⁻¹) and NO₂-N (0.031 mg L⁻¹), while February recorded high levels of NO₃-N (0.233 mg L⁻¹) (Figure 3), perhaps as a result of surface runoff and the inflow of organic and inorganic materials, as well as nutrients, from the crops and other catchment regions brought on by this season’s intense rainfall. Sewage, industrial wastes, untreated aquaculture wastes, and excess aquaculture feeds can all contribute to nutrient levels in rivers through surface runoff [41].

Additionally, May recorded significantly high average values of COD (4.16 mg L⁻¹), BOD (5.00 mg L⁻¹), and TP (0.173 mg L⁻¹), which could be attributable to various intertwined environmental processes, anthropogenic influences, and seasonal climatic variation [42]. One of the primary factors this season is elevated temperatures, which can significantly influence biological activity and increase the rate of organic matter decomposition, elevating BOD and COD levels in the lagoon. Higher temperatures reduce dissolved oxygen levels and enhance microbial metabolism, increasing oxygen consumption and a corresponding rise in BOD [11,43]. Studies have demonstrated that organic matter decomposition and intensifying algal blooms, often fueled by agricultural runoff, can lead to elevated TP, BOD, and COD concentrations in summer months [44,45]. Meanwhile, August (Summer) recorded the highest average values of ORP (129.44 mV) and TN (0.242 mg L⁻¹), a finding that could be partially linked to the elevated summer temperatures, which enhance microbial biological activity, a phenomenon that can affect ORP as increased microbial respiration and decomposition of organic matter elevate the oxidative conditions in the water [46]. Further, elevated TN levels could result from nutrient influx from various anthropogenic sources, including untreated sewage discharge and agriculture [7]. Additionally, high temperatures can accelerate mineralization of nitrogen compounds, thereby contributing to an increase in TN levels [47]. Water pressure (WP) varied significantly, with high values recorded in November (765.16 mm Hg) and February (765.36 mm Hg), and the lowest value in May (749.90 mm Hg), possibly attributable to groundwater-surface water interactions.

3.2. Phytoplankton Community Structure

This study identified 109 phytoplankton species belonging to 38 genera, and 5 phyla, including Cyanophyta, Bacillariophyta, Chlorophyta, Cryptophyta, and Dinophyta. The phytoplankton density ranged from 2.12 × 10² to 4.61 × 10⁴ Ind L⁻¹, with an average of 1.65 × 10³ Ind L⁻¹. Generally, Bacillariophyta showed the highest average contribution of 4.06 × 10³ Ind L⁻¹ (82.26%), followed by Cyanophyta 5.75 × 10² Ind L⁻¹ (11.64%), while Cryptophyta had the lowest density 6.40 Ind L⁻¹ (0.13%). This reveals that the lagoon’s phytoplankton community is strongly dominated by Bacillariophyta, which are associated with high nutrient enrichment [48]. Bacillariophyta’s tolerance to saline water could also explain their dominance, as they have hard siliceous shells and tolerate saline waters [49]. The results are consistent with wang et al. [43], who reported a high abundance of Bacillariophyta species in saline landscape waters in China. This finding also agrees with Aubry et al. [50], who also reported diatom dominance in the Gulf of Venice, Northern Adriatic Sea. Similarly, the dominance of Bacillariophyta observed here is also consistent with the findings of Zhou et al. [51]. Cyanophyta is typically an indicator of eutrophication pressure and potential algal blooms. Previous studies showed Cyanophyta have been reported to associate with increased nutrient enrichment and eutrophication in water bodies [52].

The average phytoplankton density was not significantly different between the stations (Kruskal–Wallis, H = 10.24, p = 0.175) and seasons (H = 5.18, p = 0.159). As illustrated in Figure 4, the highest phytoplankton density was recorded at station Y2 (1.26 × 10⁴ Ind L⁻¹), followed by Y5 (6.03 × 10³ Ind L⁻¹) and Y4 (5.50 × 10³ Ind L⁻¹), while lower density was found at Y6 (1.35 × 10³ Ind L⁻¹) and Y8 (1.16 × 10³ Ind L⁻¹). Overall, the spatial pattern of phytoplankton density strongly relates to the gradient of organic load and nutrient conditions in the lagoon, shaped by hydrology and local anthropogenic inputs. findings revealed spatial variations in phytoplankton density, which can be partially linked to differences in environmental conditions. Notably, Bacillariophyta was most dominant in all the stations, while remarkable density of Chlorophyta (11.07%), Dinophyta (22.16%), and Cyanophyta (58.95%) was recorded at Y1, Y6, and Y8, respectively (Figure 5a). This could be explained by variations in environmental heterogeneity across sampling sites, which affect nutrient levels, light, and water movements, resulting in differing phytoplankton densities.

As shown in Figure 5b, Bacillariophyta was the most dominant throughout all the seasons. At the same time, the highest and lowest number of phytoplankton species was recorded in May (53 species in 38 genera) and November (33 species in 28 genera). Diatoms are often the dominant phytoplankton groups in many aquatic ecosystems due to their competitive advantage in efficient nutrient uptake (especially silica, nitrate, phosphate) and photosynthetic rates under a wide range of conditions. In addition, they are adapted to thrive in varying light and temperature levels across all seasons; their silica frustules, cell walls, offer defense against some grazers and environmental stress [53]. Their high density in May can be attributed to the warmer temperatures and increased light penetration, which favor their growth. Moreover, summer storms or localized upwelling can provide a temporary nutrient boost, allowing them to bloom during this season [54]. Furthermore, the highest and lowest phytoplankton densities were recorded in November (1.13 × 10⁴ Ind L⁻¹) and February (1.84 × 10³ Ind L⁻¹). At Y6, elevated levels of COD, BOD, total phosphorus (TP), and chlorophyll-a indicate high organic matter and nutrient enrichment, creating conditions favorable for Dinophyta (dinoflagellates). Dinoflagellates often thrive in nutrient-rich, organically enriched waters and can form blooms under such conditions, which is consistent with your observations. However, the elevated organic and nutrient enrichment inside the lagoon (Y6) also raises concerns about potential harmful algal blooms or oxygen depletion linked to dinoflagellate proliferation [50]. On the contrary, Y1 was characterized by the lowest values of COD, BOD, TP, and chlorophyll-a, corresponding to a less eutrophic environment. Thus, the higher abundance of Chlorophyta here may be due to fresher, marine-influenced waters with lower organic loads, which favor green algae, often the dominant species in more oligotrophic or well-mixed sites. As significant primary producers in both aquatic and terrestrial environments, Chlorophyta have a wide range of morphological and ultrastructural traits that increase their viability and ecological niche in comparison to other phytoplankton [55]. At the same time, Cyanophyta was among the highest at the intertidal sites (Y8). Cyanobacteria can tolerate fluctuating conditions, such as intermittent exposure, higher temperatures, and stratified or shallow waters, which may lead to higher nutrient recycling.

In contrast, Bacillariophyta, the genus with the most significant percentage of diatoms, is highly viable due to its unique carbon fixation process and high concentration. According to previous studies, physiological advantages allow Bacillariophyta to persist as the dominant phytoplankton in aquatic environments across all seasons. In addition, the highest and lowest phytoplankton density was recorded in November (1.13 × 10⁴ Ind L⁻¹) and February (1.84 × 10³ Ind L⁻¹). In this study, November was characterized by temperatures and adequate light availability, which creates a favorable condition for phytoplankton growth. The autumn mixing and nutrient availability also fuel secondary phytoplankton blooms [56]. Further, in autumn, the releases of nutrient limitation due to mixed layer deepening support phytoplankton growth before returning to winter conditions [57]. Winter season, on the other hand, features deeper mixed layers and lower light availability that dilute phytoplankton concentrations and reduce net primary production [58]. Low light levels throughout the winter allow for limited photosynthesis, making the respiratory loss term correspondingly more significant, even if the area is frequently exposed to the euphotic zone and is not restricted by nutrients [58]. Several physiological and morphological factors during winter determine the makeup of the surviving phytoplankton community. It has been reported that non-growing or slowly growing cells sink more quickly than blooming cells [59]. Grazing pressure can become decoupled from phytoplankton due to these physical conditions, resulting in lower densities. However, some resilience or active growth may still occur beneath the mixed layer via convective processes.

Interestingly, One-way ANOSIM for phytoplankton community composition showed significant seasonal dissimilarity (R = 0.828, p < 0.001). Additionally, the PCA examined the phytoplankton community composition. Components 1 and 2 explained 15.5% and 12.4% of the total variation, respectively (Figure 6). Similarly, the results showed seasonal separation of phytoplankton species (Figure 6). For example, phytoplankton species in May mainly comprised Aulacoseira granulata, Ceratocorys sp., Chaetoceros sp., and Microcystis sp. Meanwhile, August consisted of Biddulphia sinensis, Coscinodiscus subtilis, Cymella tumida, Lauderia annulata, Navicula sp. Tripos gallicus, and Oscillatoria princeps. At the same time, November comprised Chlorella vulgaris, Coscinodiscus curvatulus, Thalassiosira angustelineata, Gaillonella sulcata, etc. Therefore, SIMPER results revealed that Ceratocorys sp., Chaetoceros sp., Coscinodiscus subtilis, Oscillatoria princes, and Thalassionema nitzschioides contributed to the seasonal difference in phytoplankton composition (

Table 2. SIMPER analysis results for phytoplankton species abundance and composition based on seasons.

Species	May vs. Aug		May vs. Nov		May vs. Feb		Aug vs. Nov		Aug vs. Feb		Nov vs. Feb
	Av. Dissim	Contrib. %	Av. Dissim	Contrib. %	Av. Dissim	Contrib. %	Av. Dissim	Contrib. %	Av. Dissim	Contrib. %	Av. Dissim	Contrib. %
Aulacoseira granulata	1.05	1.07	4.04	4.05	4.05	4.05	0.00	0.00	0.00	0.00	0.00	0.00
Bacillaria paxillifera	0.00	0.00	0.04	0.04	0.00	0.00	0.01	0.01	0.00	0.00	6.25	6.39
Biddulphia obtusa	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.65	0.67
Biddulphia sinensis	5.77	5.88	0.03	0.03	0.03	0.03	9.79	9.79	9.82	9.82	0.00	0.00
Syringidium daemon	0.30	0.31	0.93	0.93	0.92	0.92	0.19	0.19	0.19	0.19	8.02	8.21
Ceratocorys sp.	4.47	4.56	18.40	18.44	18.47	18.48	0.19	0.19	0.19	0.19	0.00	0.00
Chaetoceros decipiens	0.00	0.00	0.01	0.01	0.00	0.00	0.00	0.00	0.00	0.00	2.09	2.14
Chaetoceros sp.	7.61	7.76	27.72	27.78	27.81	27.83	0.19	0.19	0.19	0.19	1.78	1.82
Chlorella vulgaris	0.05	0.05	0.35	0.35	0.35	0.35	0.00	0.00	0.00	0.00	1.98	2.02
Coscinodiscus curvatulus	0.35	0.35	0.03	0.03	0.00	0.00	0.41	0.41	0.40	0.40	10.11	10.35
Coscinodiscus sp.	0.38	0.39	0.20	0.20	0.20	0.20	0.60	0.60	0.60	0.60	1.46	1.49
Coscinodiscus subtilis	15.11	15.39	0.00	0.00	0.00	0.00	16.85	16.85	16.86	16.86	0.00	0.00
Thalassiosira angustelineata	0.00	0.00	0.07	0.08	0.00	0.00	0.02	0.02	0.00	0.00	3.09	3.17
Cymella tumida	1.93	1.97	0.00	0.00	0.00	0.00	3.49	3.49	3.50	3.50	0.33	0.33
Neomoelleria cornuta	0.00	0.00	0.20	0.20	0.00	0.00	0.06	0.06	0.00	0.00	9.19	9.40
Eucampia zoodiacus	0.00	0.00	0.00	0.00	0.01	0.01	0.00	0.00	0.00	0.00	3.87	3.97
Gomphonema sp.	0.54	0.55	0.12	0.12	0.10	0.10	0.79	0.79	0.79	0.79	2.33	2.38
Lauderia annulata	2.38	2.42	0.00	0.00	0.00	0.00	4.19	4.20	4.21	4.21	0.00	0.00
Leibleinia gracilis	1.28	1.30	9.50	9.52	9.59	9.60	0.01	0.01	0.00	0.00	13.08	13.39
Lyngbya sp.	0.00	0.00	0.05	0.05	0.00	0.00	0.01	0.01	0.00	0.00	6.00	6.14
Melosira inflexa	0.53	0.54	4.34	4.34	4.40	4.40	0.00	0.00	0.01	0.01	11.37	11.63
Gaillonella sulcata	0.00	0.00	0.01	0.01	0.00	0.00	0.00	0.00	0.00	0.00	3.02	3.09
Merismopedia tranquilla	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00
Microcystis sp.	1.79	1.82	6.18	6.19	6.20	6.20	0.00	0.00	0.00	0.00	0.00	0.00
Navicula sp.	1.11	1.13	7.13	7.14	7.25	7.25	2.02	2.02	2.04	2.04	9.42	9.65
Tripos gallicus	2.43	2.48	0.00	0.00	0.00	0.00	4.40	4.40	4.41	4.41	0.00	0.00
Oscillatoria princeps	10.28	10.48	0.00	0.00	0.00	0.00	13.33	13.34	13.35	13.35	0.00	0.00
Oscillatoria sp.	1.58	1.61	9.98	10.00	10.05	10.05	0.00	0.00	0.00	0.00	2.76	2.82
Phormidium spirale	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.65	0.67
Skeletonema costatum	2.77	2.82	10.47	10.49	10.51	10.51	0.00	0.00	0.00	0.00	0.00	0.00
Thalassionema nitzschioides	34.95	35.61	0.00	0.00	0.00	0.00	40.68	40.70	40.71	40.72	0.26	0.27
Thalassiothrix frauenfeldii	1.49	1.52	0.00	0.00	0.00	0.00	2.71	2.71	2.72	2.72	0.00	0.00
Overall average dissimilarity (%)	98.15		99.80		99.96		99.95		100.00		97.91

Note: Significant contributions to dissimilarities are in bold.

). For example, Ceratocorys sp. contributed 18.44% and 18.48% of the dissimilarities between May and November (May–Nov) and May and February (May–Feb), respectively (Table 2). Similarly, Chaetoceros sp. contributed 27.78% and 27.83% of the dissimilarities between May-Nov and May-Feb. Coscinodiscus subtilis contributed 15.39%, 16.85%, and 16.86% of the dissimilarities between May-Aug, Aug-Nov, and Aug-Feb, respectively. At the same time, Oscillatoria princeps contributed 10.48%, 13.34%, and 13.35% of the dissimilarities between May and August, August and November, and August and February, respectively. In addition, Thalassionema nitzschioides contributed 35.61%, 40.70%, and 40.72% of the dissimilarities between May and August, August and November, and August and February, respectively.

3.3. Dominance Characteristics of Phytoplankton Community

According to the dominance calculations (Y > 0.02), there were 30 dominant phytoplankton species, of which 7, 22, 2, and 1 belonged to Cyanophyta, Bacillariophyta, Dinophyta, and Chlorophyta, respectively (Table 2). Notably, Oscillatoria sp. and Navicula sp. were the dominant phytoplankton species in all the seasons, except August, contributing significantly to the total phytoplankton density. Oscillatoria sp. and Navicula sp. are often reported as dominant phytoplankton species across seasons in various aquatic environments. The dominance of Oscillatoria sp. has been linked to anthropogenic nutrient input and correlated with specific nutrient concentrations, such as phosphate, which support persistent blooms. Similarly, Gao et al. [60] reported Oscillatoria and Navicula among the dominant algal taxa in the Yangtze River, with their community structure influenced by temperature, nutrients, and salinity. In addition, Thalassiothrix frauenfeldii, Thalassionema nitzschioides, Lauderia annulata, Biddulphia sinensis, Cymella tumida, Oscillatoria princes, Neoceratium deflexum, and Coscinodiscus subtilis were dominant during August. Thalassiothrix frauenfeldii and Thalassionema nitzschioides are known to exhibit seasonal dominance, which is linked to variations in temperature, salinity, and nutrient availability. For instance, in waters around Macau, Thalassiothrix frauenfeldii and Thalassionema nitzschioides populations form part of a community detected in late summer and early autumn, indicative of dominance during warmer months like August [39]. Biddulphia sinensis and Cymella tumida have also been reported to appear during warmer seasons, with August often corresponding to rainy or post-monsoon periods that favor their growth [61]. Meanwhile, Chaetoceros decipiens, Cerataulina daemon, Gomphonema sp., Eucampia cornuta, Lyngbya sp., Melosira sulcata, Coscinosira polychorda, Chlorella vulgaris, Coscinodiscus curvatulus, and Bacillaris paradoxa were dominant in November. At the same time, Melosira moniliformis and Chaetoceros sp. were more dominant during May and February (see

Table 3. Dominant species and dominance of phytoplankton.

Phyla	Dominant Species (Y ≥ 0.02)	May	Aug	Nov	Feb
Cyanophyta	Microcystis sp.	0.06	-	-	-
	Leibleinia gracilis	0.08	-	0.17	-
	Oscillatoria sp.	0.08	-	0.02	0.07
	Oscillatoria princeps	-	0.11
	Lyngbya sp.	-	-	0.07
	Merismopedia tranquilla	-	-	-	0.05
	Phormidium spirale	-	-	-	0.04
Bacillariophyta	Aulacoseira granulata	0.04	-	-	-
	Melosira inflexa	0.03	-	-	0.27
	Gaillonella sulcata	-	-	0.03	-
	Thalassiosira angustelineata	-	-	0.03	-
	Thalassiothrix frauenfeldii	-	0.02	-	-
	Thalassionema nitzschioides	-	0.38	-	-
	Lauderia annulata	-	0.03	-	-
	Biddulphia sinensis	-	0.07	-	-
	Cymella tumida	-	0.02	-	0.02
	Coscinodiscus subtilis	-	0.16	-	-
	Skeletonema costatum	0.10	-	-	-
	Navicula sp.	0.05	-	0.10	0.08
	Chaetoceros sp.	0.26	-	-	0.06
	Chaetoceros decipiens	-	-	0.03	-
	Cerataulina bicornis	-	-	0.11	-
	Gomphonema sp.	-	-	0.02	-
	Neomoelleria cornuta	-	-	0.09	-
	Coscinodiscus curvatulus	-	-	0.11	-
	Bacillaria paxillifera	-	-	0.06	0.02
	Eucampia zoodiacus	-	-	-	0.10
	Biddulphia obtusa	-	-	-	0.05
	Coscinodiscus sp.	-	-	-	0.04
Dinophyta	Ceratocorys sp.	0.17	-	-	-
	Tripos gallicus	-	0.03	-	-
Chlorophyta	Chlorella vulgaris	-	-	0.03	-

).

3.4. Phytoplankton Diversity

Phytoplankton diversity can reflect the status of water quality; for example, a high diversity value may indicate unpolluted water, whereas low diversity suggests polluted water [7,62]. The Shannon, Margalef, and Evenness indices in Xiaohai lagoon ranged from 0.19 to 2.24, 1.11 to 4.25, and 0.07 to 0.87, averaging 1.53, 2.48, and 0.60, respectively, indicating moderate pollution [63]. There was an insignificant spatial difference between the sampling stations regarding Shannon diversity (Kruskal–Wallis, H = 6.59, p = 0.473), Margalef Richness (H = 3.34, p = 0.852), and Pielou Evenness (H = 8.36, p = 0.302). As demonstrated in the spatial distribution maps (Figure 7a–d), the values of Shannon, Margalef, and Evenness indices were highest at the lagoon mouth (Y1), while decreasing inside the lagoon, recording lower values at Y4 and Y5. Furthermore, the Margalef Richness showed significant seasonal variation (H = 14.04, p = 0.003), recording the highest and lowest values in August and November, respectively (Figure 7d). Meanwhile, there was insignificant seasonal variation in Shannon diversity (H = 1.90, p = 0.593) and the Evenness Richness index (H = 1.88, p = 0.597). Usually, the J′ (evenness) ranges from 0 to 1, with J′ > 0.3 considered an excellent standard of phytoplankton diversity [64].

3.5. Influence of Environmental Factors on Phytoplankton Community

The greatest axis length of 6.02, exceeding 3.0, was observed when the phytoplankton community was analyzed using DCA. Therefore, the unimodal ordination approach, CCA, investigated how environmental conditions drive the organization of phytoplankton communities. The variation was explained by the CCA 1 and CCA 2 axes in 20.1% and 17.1% of cases, respectively (Figure 8), revealing strong seasonal niche partitioning driven by distinct environmental variables. During August, the phytoplankton species, such as Biddulphia sinensis, Coscinodiscus curvatulus, Coscinodiscus sp., Coscinodiscus subtilis, Cymella tumida, Gomphonema sp., Lauderia annulata, Navicula sp., Tripos gallicus, Thalassionema nitzschioides, and Thalassionema frauenfeldii, were strongly associated with TN, TDS, salinity, ORP, EC, and WT. This assemblage reflects adaptation to the elevated salinity, ionic strength, redox conditions, and thermal regime characteristic of late summer, underscoring the role of nutrient enrichment and physicochemical gradients in shaping community composition during this period. The influence of nutrient concentration on phytoplankton communities was illustrated by the positive relationship found between the abundance of certain phytoplankton species and nutritional elements (TN) [65]. These species are therefore useful indicators of eutrophication, organic contamination, and nutrient enrichment. Additionally, nutrients play a crucial role in controlling phytoplankton growth and reproduction; earlier studies have shown that higher nutrient levels promote eutrophication [66,67].

Wang et al. [43] reported that fertilizers mainly affected the dispersion of phytoplankton communities in Chinese landscape waterways, as compared to earlier research. Furthermore, the phytoplankton community structure in Shengjin Lake was found to be highly impacted by turbidity, conductivity, and nutrients [68]. According to a related study by Sun et al. [69], the primary variables affecting the phytoplankton community were nitrate nitrogen, water temperature, total nitrogen, and total suspended particles. Other studies have shown a positive correlation between nutrient concentration and phytoplankton abundance [70,71]. As previously mentioned, nutrients play a significant role in controlling the growth and reproduction of phytoplankton. Numerous investigations have shown that the limiting variables of phosphorus and nitrogen are linked to eutrophication [66,67]. For instance, N and P enrichment may shift the phytoplankton community from one dominated by diatoms to one dominated by cyanobacteria [72]. Meanwhile, during May, Aulacoseira granulata, Syringidium daemon, Ceratocorys sp., Chaetoceros sp., Melosira inflexa, Microcystis sp., Oscillatoria sp., and Skeletonema costatum were positively associated with TP and COD. TP acts as a key limiting nutrient, stimulating the growth of high-biomass bloom formers. The presence of Microcystis and Oscillatoria, in particular, is a classic bio-indicator of phosphorus loading, often from agricultural runoff or organic waste. High COD signifies a substantial load of oxidizable organic matter, consistent with fertilizer runoff, aquaculture effluent, or terrestrial organic input. This organic matter fuels microbial respiration, which can contribute to bottom-water hypoxia, subsequently promoting phosphorus release from sediments (internal loading), a positive feedback loop that further exacerbates eutrophication. The co-occurrence of diatoms and cyanobacteria in May suggests a transitional phase where pulsed nutrient inputs simultaneously fuel the rapid growth of diatoms (Skeletonema, Chaetoceros) and create conditions (potential hypoxia, high nutrient loads) that favor the emergence of later-summer cyanobacterial blooms. This pattern aligns with global observations in eutrophic estuaries, where spring nutrient pulses initiate a succession towards noxious cyanobacteria dominance [73].

Additionally, Bacillaria paxillifera, Chaetoceros decipiens, Neomoelleria cornuta, Lyngbya sp., and Gaillonella sulcata were positively associated with NH₃-N, NO₃-N, DO, and WP during November (Figure 8). The association of these phytoplankton species with NH₃-N and NO₃-N suggest increased allochthonous input, likely from surface runoff and riverine discharge following rainfall events. This nutrient surge fuels the growth of diatoms like Gaillonella sulcata, which are known to thrive in high-nutrient, turbulent environments. High DO concentrations and the correlation with WP suggest improved vertical mixing and oxygenation. This disrupts thermal stratification, suppressing buoyant cyanobacteria (common in summer) and favoring non-motile diatoms that benefit from a well-mixed water column. The presence of Lyngbya sp., a filamentous, often benthic cyanobacterium, alongside diatoms suggests a community adapted to nutrient-rich, shallower, or perturbed habitats. Its association with NO₃-N aligns with its known preference for bioavailable nitrogen sources. Generally, this assemblage represents a transitional community, benefiting from nutrient influx but constrained by physical mixing, which sets the stage for the subsequent winter community structure.

4. Conclusions

This study examined the environmental factors and seasonal succession of phytoplankton communities in Xiaohai Lagoon, Hainan Island, China. The findings unravel the key environmental drivers controlling phytoplankton community dynamics in the Xiaohai Lagoon. While phytoplankton abundance remained relatively stable spatio-temporally, the community composition was highly responsive to nutrient gradients and physicochemical conditions, with Total Phosphorus (TP), Total Nitrogen (TN), and salinity emerging as the most significant factors. Physicochemical parameters, including TP, COD, TN, TDS, salinity, ORP, EC, and WT, significantly influenced the species composition and abundance. This indicates that the lagoon’s ecosystem health is more accurately reflected by shifts in species identity rather than total cell count. Secondly, the phytoplankton community was dominated by diatoms (Bacillariophyta), which suggests that the lagoon, despite its nutrient-rich environment, is currently in a state that favors this functional group. The significant spatial differences in water quality parameters, coupled with the seasonal pattern of higher values in summer, highlight the combined pressure of anthropogenic nutrient inputs and natural seasonal cycles. Overall, the research offers critical insights for managing Xiaohai Lagoon, highlighting strategies that address spatio-temporal variations in water quality and phytoplankton dynamics.