The sampling was collected at 34 sites throughout the Guangdong mangrove forest, at latitude 20°12′—25°31′N and longitude 109°45′—117°20′E (Fig. 1). The sampling sites are characterized by different species of true mangroves, such as: Acrostichum aureum, Excoecaria agallocha, Bruguiera gymnorrhiza, Kandelia obovata, Rizhophora stylosa, Lumnizera racemosa, Aegicera corniculatum, Avicennia marina and Acanthus ilicifolius, and non-true mangrove species e.g., Pongamia pinnata, Hibiscus tiliaceus, Thespesia populnes, Herltiera littoralis, Cerbera manghas, Clerodendrum inerme and Pluchea indica (Inyang and Wang, 2020). The mangrove ecosystems in Guangdong are influenced by semidiurnal to mixed tides all year long, with a tidal range of 0.8 m during neap tides and 4.0 m during spring tides. The average annual precipitation in the study region is ~2,000 mm (Liu, 2013), and approximately 85% of the total rainfall is concentrated in the rainy season from April to September. The study sites were selected along the mangrove ecosystem of the Guangdong province, China, from the Leizhou Peninsula in the west to the east. Some sections of the mangrove floor are covered with algal bloom (Fig. 2), which has resulted in the death of some mangrove species (Fig. 2a).

Water quality monitoring was carried out in the dry season from October, 2017 to January, 2018, and in the wet season from July to August, 2018. At each investigation site, the environmental parameters such as water temperature, pH, salinity, electrical conductivity (EC), turbidity and total dissolved solids (TDS) were measured in-situ using a Quanta® Water Quality Monitoring System (Hydrolab Corporation, USA) during high water level within the depth of 1m. 300 mL of surface water was sampled using GO FLO bottles for analysis of nutrients (phosphate, nitrate, nitrite and silicate). The water sample was filtered using 0.22 μm Whatman® GF/F filters before analysis. Within 2-3 h after sampling, nitrate (NO₃-N), nitrite (NO₂-N) and silicate (SiO₃-Si) were analyzed with a SKALAR auto-analyzer (Skalar Analytical B.V. SanPlus, Holland). Phosphate (PO₄-P) was pretreated using an oxidization method by hypobromite before analysis using an auto-analyzer. Chlorophyll a analysis was carried out according to HELCOM (2015) using a SHIMADZU Spectrophotometer (UV-1700, Japan).

Multivariate analytical techniques were employed to extract the spatiotemporal patterns and trends in water quality data to enhance the understanding of hidden information within the data. Various multivariate statistical methods were used in our study, including correlation analysis, principal component analysis and cluster analysis, using PRIMER v.7 (PRIMER-E Ltd, Roborough, Plymouth, UK).

The trophic state was evaluated by single factor index, trophic state index (TSI) Chl a. The equation for TSI (Chl a) is as follows (Wang et al, 2002):

Evaluation standard: 0<TSI≤30 oligotrophic, 30<TSI≤ 50 mesotrophic, TSI>50 eutrophic, >70 high eutrophic.

The environmental variables of the mangrove ecosystem along the Guangdong coast, South China Sea are presented in Table 1. Meanwhile, various levels of significance (P<0.01 or P<0.05) were established among the measured physiochemical parameters (Tab. 2). Generally, the nitrate concentration across the mangrove forest fluctuated greatly in the dry season than in the wet season. Generally, the spatiotemporal variation of the nutrient salts was significantly different between the two seasons (P<0.01), with the dry season recording the high concentration values (Tab. 1). Although there were spatial differences in nutrients distribution between the left and right sides of the Leizhou Peninsula (Fig. 3), the mean nutrient concentration across the left side was significantly lower than the east part of the Guangdong coast during both seasons (P<0.01). Meanwhile, nutrient salts concentration showed no significant difference (P<0.01 or P<0.05) between the two sides of the Leizhou Peninsula. Chl a concentrations varied greatly across the sampling sites with a mean value of 1.82 mg·m⁻³ and 8.38 mg·m⁻³ in both the wet and dry season, respectively; and were influenced by pH, salinity, and nutrient salts (Tab. 2). Its highest value in the wet season was 13.53 mg·m^-3 (GD11), while the lowest value was 0.14 mg·m⁻³ (GD20), while in the dry season, its maximum and minimum values were recorded as 0.75 mg·m⁻³ (GD23) and 38.08 mg·m⁻³ (GD36), respectively. Moreover, the Chl a concentration was significantly higher on the east coast than the left side of the Leizhou Peninsula (P<0.01). Temporally, Chl a concentration varied significantly between the dry and wet seasons (P<0.05), but showed no significant difference between the left and the right side of the Leizhou Peninsula (P<0.01), Fig. 3.

The Correlation coefficient was used to confirm the linear relationship between the environmental variable and the TSI (Chl a) and algal cells abundance during both seasons (Tab. 2). TSI (Chl a) evaluation indicated that trophic states of the study area ranged from oligotrophic to eutrophic (Fig. 2). There was a significant difference between the rainy season and the dry season for the TSI (P<0.05). The mean TSI in the wet season was 25±11.2, which was significantly lower than that in the dry season (mean TSI 43±10.8) (P<0.05). At the Leizhou Peninsula, the TSI (Chl a) showed no significant difference between the left and right side of the Leizhou Peninsula during both seasons but spatial differences were found. Relatively, the left side of the Leizhou Peninsula reported high TSI values than the right side and was significantly different from the east part of the coast (Fig. 4). Meanwhile, across the mangrove forest, in the dry season, the eutrophication state was alarming than in the wet season. Based on environmental variables pattern and TSI, we found that hydrological variables such as; pH, nitrate, and nitrite were a vital component of the ecological process that influenced the trophic state in the mangrove ecosystem (P<0.01 or P<0.05) during the dry season. Other parameters that positively impacted the TSI were salinity and EC during both seasons. The algal cell abundance in the mangroves was greatly influenced by nitrate and nitrite (P<0.01 or P<0.05) during the dry season, while in the rainy season, a positive relation was observed with phosphate and salinity (Tab. 2). Its spatiotemporal distribution (Fig. 5) showed that the maximum algal density was recorded as 38.97×10⁶ cells·L^-1 (GD15) and as 13.17×10⁶ cells·L^-1 (GD40) in the wet and dry seasons, respectively. The algal density was higher in the wet season than in the dry season (P<0.05, Fig. 4). Moreover, its spatial distribution was significantly different during the wet and dry seasons (P<0.01).

Water temperature is influenced by weather conditions of an area or region. Still, it could also be affected by the industrial discharge of thermal effluence within a local niche (Jiang et al, 2013), causing biological imbalance in the ecosystem. Salinity is known to be an important parameter in the estuarine and mangroves that affects the distribution and physiology of living fauna and flora. The spatiotemporal variability in the salinity levels in the study area was within the range of brackish water environment (2.15‰~28.82‰). Similarly, observations were made at different mangrove ecosystems during the dry and wet season (Miyittah et al, 2020). These studies attributed the low salinity content in the surface water to strong rainfall events and an influx of freshwater into the system. Across the mangroves, a temporary increase in salinity content coincided with a decrease in water temperature and vice versa. This explains why evaporation plays little or no role in salinity variation; rather, marine/freshwater exchange coupled with surface water dilution by rainfall was the key player in salinity variation. Salinity is a limiting factor in the distribution of living organisms, and its changes are likely to affect biological communities (Inyang and Wang, 2020). The pH in surface waters remained alkaline throughout the study period, and its range value was within the limit reported elsewhere in the mangroves (Toriman et al, 2013; Miyittah et al, 2020; Samara et al, 2020). Generally, pH plays a vital role in the distribution of biological communities in the estuaries (Inyang and Wang, 2020). The EC variation in surface waters is influenced by dissolved ions of natural and anthropogenic origins. Inyang et al (2018) concluded that EC significantly depends on salinity, TDS, total cation and anion concentration in the mangrove ecosystem. Higher EC in the dry season could be attributed to the influx of saline water plus contaminated substances from the fish farming facilities, while its high value during the wet season is an indication that a proportion of the dissolved ions were transported into the forest by surface runoff from non-point sources such as agricultural farmlands and fish farming facilities which are dominant land use within the littoral fringes.

Turbidity measurement in surface waters is an indication of the presence of suspended solids (plankton, sediment particles, etc.) and dissolved organic matter. The observed positive correlation between turbidity and Chl a during both seasons confirmed that particles in suspension in the mangrove forest are nutrient-rich that enhance the productivity of phytoplankton. Inyang et al (2018) concluded that rainfall directly increased the turbidity load through runoff. Turbidity is the most visible indicator of water quality in coastal waters. Increased turbidity can hinder sunlight penetration in coastal surface waters, thereby affecting phytoplankton productivity and vice versa. The right side of the Leizhou Peninsula was observed to have experienced high turbidity during both seasons due to the large volume of waste water discharged from the fish pond facilities, small fragmented mangrove areas that have little buffering capacity and tidal influence. The significant relationship between turbidity and salinity indicates that salinity is largely influenced by the processes that increased the turbidity level in the mangrove ecosystem. The nutrient salts fluctuate spatially and temporally across the mangrove ecosystem. The coastal waters in China are known to be eutrophic due to nutrient generating activities such as aquaculture, agriculture, industrial activities, sewage disposal from the urban cities and tourism along the coast. The temporal variation in the nutrients may have been resulted from biological activities, physical and chemical properties of the water. Meanwhile, the changes in nutrient concentration in this study could be attributed to the difference in catchment characteristics and mangrove coverage area. MacDonnell et al (2017) reported that nutrient salts were indicators of the mangroves’ hydrogeomorphic settings, including tidal fluxes and urban storm water runoff. High nutrient concentration in the mangroves could generally lead to accelerated changes (Geedicke et al, 2018), such as alteration of biological processes and increased N2O fluxes to the atmosphere, which may contribute to global warming (Reis et al, 2017) and reduction in services such as improvement of water quality (Reis et al, 2017). Tjandraatmadja and Diaper (2006) confirmed that the increase in nutrient concentration in the surface water essentially comes from municipal wastes and runoff from agricultural effluents.

Chl a concentration that reflects the primary productivity and trophic condition in estuaries, coastal and oceanic waters is the net result (standing stock) of both growth and loss processes. Its positive corrections with nitrate, nitrite and phosphate explain the effect of the nutrient salts on the water quality of the mangroves. Manju et al (2012) establish a positive correlation between nutrient salts and Chl a in the mangroves of the Kerala Coast. Generally, high nutrient concentration in coastal waters causes eutrophication (Song et al, 2009; Hou et al, 2018). The study established a positive and significant temporal difference (P<0.05 or P<0.01) between nutrient salts and the trophic state index. The trophic state analysis confirmed that the surface water of the mangrove ecosystem during the dry season was mainly mesotrophic to eutrophic, whereas in the wet season it was eutrophic, mesotrophic, and oligotrophic. The spatiotemporal difference in trophic index variation from the left of the Leizhou Peninsula to the east part of the coast could be attributed to differences in catchment characteristics that possibly affect nutrient concentrations and turbidity levels; changes in pH, salinity and EC that affect algal development; and changes in mangrove coverage areas that provide a calm and enabling environment for algal proliferation. The coastal eutrophication in the mangroves has been linked to the impact of anthropogenic nutrient input (Valiela et al, 2009). This nutrient input, together with other factors (pH, salinity, EC, etc.) and zooplankton grazing, causes Chl a concentration to be either high or low in the mangroves.

Cluster analysis which detected similar groups between the sampling sites in the mangroves during the dry and wet season has significantly delineated 3 groups and 2 groups of samples during the dry and wet season, respectively. These sample groups were impacted by different environmental parameters, mostly by turbidity, EC, salinity, water temperature and nutrient salts. Turbidity and nutrient salts concentration are associated with human activities along the littoral fringes of the mangroves and therefore serve as a better environmental indicative parameter for future monitoring. The trophic differences in the study area could be linked to the anthropogenic nutrient input (Valiela et al, 2009) and mangrove hydrogeomorphic differences (MacDonnell et al, 2017). The CA revealed that the water quality during the wet season was relatively better compared to the dry season. In contrast, MacDonnell et al (2017) reported high nutrient concentrations during the wet season at mangrove wetlands in Southwest Florida. It can be stated that hierarchical CA provided a reliable tool for surface water analysis in the study area (Inyang et al, 2018) and made it possible to optimize future monitoring programs. The PCA results explain that nutrient salts play an important role in the water quality across the Guangdong mangroves. Though other physiochemical parameters exerted significant change on-site, their effects on the algal community cannot be overemphasized (Inyang and Wang, 2020). Human activities have a strong influence on the mangroves in the Guangdong coastal area.