This study investigates the relationship between macrobenthic functional group composition and hypoxia in the Changjiang River estuary and its adjacent sea areas. A total of 82 stations were divided into three areas, named non-hypoxic area, hypoxic area and the Changjiang River, respectively. A total of 256 macrobenthic species were collected, which were divided into five functional groups, including carnivorous, detritivorous, omnivorous, phytophagous, and the planktivorous functional group. A similarity analysis performed on the macrobenthic functional groups of the three zones indicates that the functional group distributions of non-hypoxic and hypoxic zones are not significantly different, whereas the functional group compositions of the Changjiang River estuary and the other two zones are rather different. The results of canonical correspondence analysis reveal that the distribution of macrobenthic functional groups is compounded by various environmental factors, of which dissolved inorganic nitrogen, salinity, and temperature exert a significant effect. Combining the results of previous studies, we speculate that macrobenthos are able to adapt to the occurrence of hypoxia by changing their body morphology, distribution location, and community structure. When the environmental conditions prevent the hypoxia from occurring, persisitent hypoxic zones can recover their marine microbenthic community.
Dissolved oxygen (DO) in seawater is an important parameter of the source element for organisms and is essential for the metabolism of most marine organisms. DO levels are an important indicator of the growth status of organisms and the pollution state of the environment (Zhang et al., 2007). In many sea areas, under the effect of various chemical, physical, and biological processes, hypoxic areas formed (Diaz and Rosenberg, 2008; Zhang et al., 2010). According to a 2011 report, there are more than 500 anoxic zones in the world’s oceans (Conley et al., 2011), e.g. the Gulf of Mexico (Rabalais et al., 2007), the Black Sea, the Baltic Sea (Conley et al., 2009), and the Changjiang River estuary in China (Li et al., 2002). The problem of hypoxia in sea areas will potentially become a major global ecological issue. For example, during summer, a hypoxic area with ≤2 mg l−1 of bottom water DO in the sea area of the Changjiang River estuary increased from 1,800 km2 in 1959 (Data on the comprehensive national oceanographic survey, 1961) to 19,600 km2 in 2006, including an area spanning from the sea area on the north side of the Changjiang River estuary to the central and southern sea areas of Zhejiang Province, comprising double hypoxic centres (Zhou, 2010). According to Diaz's statistical analysis of the hypoxic phenomena in more than 40 global sea areas, it is believed that eutrophication is the primary cause of the increasing severity of offshore hypoxia phenomena (Diaz, 2001), whereas spring thermocline formation hindering the water exchange between surface and bottom waters is another key factor (Zhou, 2010).
Generally, areas with DO concentrations below 2 mg l−1 are regarded as anoxic zones (Dauer et al., 1992). When DO concentrations are between 2.86 mg l−1 and 2.00 mg l−1, benthos grows slowly, migrate passively, and have interrupted life cycles (Tyson and Pearson, 1991; Rabalais et al., 2002; Sturdivant et al., 2014). Thus, there are two hypoxic area categories: hypoxia (2.86–2.00 mg l−1) and anoxic (≤2.00 mg l−1).
Eutrophication is frequent in the Changjiang River estuary (Zhang et al., 1999), which includes a large hypoxic area surrounded by many well-known fishing grounds such as Zhoushan, Changjiangkou, Lvsi, and Dasha, among others. Macrobenthos play an important role as consumers and transporters of material circulation and energy flow in marine ecosystems (Conlan et al., 2008), and because of their relatively stable living environment and long life cycles, they indicate changes in their living environment rather accurately. Relative to species composition, functional group classification can reflect habitat gradient changes and ecosystem functions more effectively (Gaudêncio and Cabral, 2007) and has been widely applied in studies on macrobenthos worldwide (Andersen, 1995; Brown et al., 2000; You et al., 2011). In the past, investigations on macrobenthos in the Changjiang River estuary have primarily focused on comparisons of the macrobenthic community structure of the estuary and traditional areas (Li et al., 2007; Wang et al., 2009), whereas comparisons between the hypoxic and non-hypoxic sea areas have not been addressed, and studies on hypoxic sea areas lack any analysis of macrobenthic functional groups (Wang et al., 2008). Therefore, in this study, we investigated the composition of macrobenthos in the Changjiang River estuary and its adjacent sea areas, and their relationship with hypoxia, to provide implications on the impact of hypoxia on marine ecosystems, as well as scientific data for the protection of the coastal marine ecology.
Materials and methods
Research sea areas and stations
Extending 6,200 km and including a drainage area of 18,106 km2, the Changjiang River is the largest river in China, whose estuary on China’s eastern coast delivers large amounts of fresh water (924,109 m3) and sediments (486,106 t) to the East China Sea each year (Rabouille et al., 2008).
In this study, according to differences in environmental factors, primarily DO, the study area was divided into three zones: a non-hypoxic zone (RZ), a hypoxic zone (AZ), and the Changjiang River estuary zone (EZ).
Macrobenthic samples were collected in August 2006 from 82 stations (Fig. 1). Maritime survey and monitoring were carried out according to Specifications for Oceanographic Survey (GB/T12763-2007) and The Specification for Marine Monitoring (GB17378-2007), in which a 0.1-m2 Van Veen grab type sediment sampler was used. Sampling was performed twice at each station, with samples sorted using a 0.5-mm mesh screen. Biological samples were collected, fixed with 5% neutral formalin, and transported to the laboratory for classifying, counting, and weighing. The water environment at the bottom of each sampling station was also measured. The primary parameters included: Dissolved oxygen(DO), temperature (T), salinity (S), PH, dissolved inorganic nitrogen (DIN), and dissolved inorganic phosphorus (DIP).
Functional group classification
According to feeding characteristics and the literature (Navarro-Barranco et al., 2013), macrobenthos were classified into five functional groups: planktophagous (Pl), phytophagous (Ph), carnivorous (C), omnivorous (O), and detritivorous (D).
Data processing and analysis
A non-parametric analysis of variance (ANOVA, Kruskal-Wallis test) was performed on the statistical data of different zones using SPSS 11.0 software, and the differences were tested. The significance level of the statistical difference was set to α = 0.05. Similarity analysis and cluster analysis were performed on the data of different zones using PRIMER 6 software. The relationships between macrobenthos and environmental parameters were analysed through canonical correspondence analysis (CCA) using Canoco for Windows 4.53 (ter Braak and Šmilauer, 1998).
Bottom water environment
The bottom water environments of the Changjiang River estuary and adjacent sea areas were complex and variable. At the survey area sites, all environmental parameters other than DO were not significantly different between the hypoxic and non-hypoxic zones. The DO concentration of the hypoxic zone was above 2.02 mg l−1 (Table 1). The water environment of the Changjiang River estuary area was quite different from that of other areas, with very low salinity but high nitrogen and phosphorus contents according to the results (Table 1).
Composition of benthic functional groups
In the survey, a total of 256 species were collected, of which 101 were carnivorous, represented by Glycera chirori, Nassarius variciferus, Lumbrineris nagae, and Odontamblyopus rubicundus; 63 species were detritivorous, including Ampelisca cyclops, Acaudina molpadioides, Phascolosoma esculenta, etc.; 20 species were omnivorous, including Bullacta exarata, Raphidopus ciliatus, Leiochrides australis, etc.; eight species were phytophagous, including Goniada japonica, Solenocera crassicornis, etc.; and 64 species were planktophagous, represented by Oratosquilla oratoria, Moerella jedoensis, Heterocyathus sp., etc.
Functional group composition and regional differences
Table 2 shows that the bio-density of the phytophagous functional group was extremely low in the non-hypoxic zone and completely absent in the hypoxic zone and Changjiang River estuary. Two major functional groups (carnivorous and detritivorous) composed a large proportion of macrobenthic species in the non-hypoxic and hypoxic zones. In the non-hypoxic zone, the carnivorous group comprised 32.8% of the total density, and the detritivorous group comprised an even a larger proportion (33.6%). Similarly, in the hypoxic zone, the proportions of carnivorous and detritivorous groups comprised 33.6% and 48.0% of the total species, respectively. However, the bio-density in the Changjiang River estuary was generally low. The proportion of the carnivorous group was as high as 61.9%, whereas that of the detritivorous functional group was low.
The distribution of biomass was similar to that of bio-density. But, the planktophagous functional group’s individual collected in the hypoxic area was greater, such as Scapharca sp., the biomass proportion of planktophagous individuals was as high as 73.3%. Similarly, the biomass proportion of the omnivorous group in Changjiang River estuary was 53.3%.
Table 3. The bio-densities of the benthic functional groups of three different zones were compared using the PRIMER 6 software. The results of one way ANOSIM revealed that the functional group distribution of the Changjiang River estuary zone (EZ) was significantly different from that of the hypoxic zone (AZ) and non-hypoxic zone (RZ), whereas in sea areas, the functional group distribution of the RZ was very similar to that of the AZ (R=-0.014, P = 0.535).
The result of non-metric multidimensional scaling (NMDS) analysis also confirmed this result. The Changjiang River estuary zone stations were scattered at the outermost periphery of the core region, whereas those in the EZ and AZ were clustered, exhibiting no significant difference (Fig. 2).
Relationship between functional group composition and environmental factors
After synthesizing the above functional group data and environmental data, we adopted a canonical correspondence analysis to examine the relationship between functional group composition and environmental factors (Figure 3). Because environmental factors in the three zones varied remarkably, the zones were examined separately.
Results from the Changjiang River estuary zone indicated that DO was not the most important factor affecting the bio-density distribution of functional groups in this zone; rather, dissolved nitrogen, salinity, and pH were the most important environmental factors. In the RZ, the correlation of DO to the bio-density distribution was lower than that of dissolved nitrogen, pH, or temperature (T). Data from the AZ stations indicated that the influence of DO was even lower, ranking fifth among the six environmental factors, whereas T and salinity (S) were important factors affecting the density of benthic functional group.
Hypoxic zones in the Changjiang River estuary and its adjacent sea areas
In the Changjiang River estuary and adjacent sea areas, fisheries of varying sizes such as Zhoushan represent China’s traditional fishing grounds. In recent years, the Changjiang River has discharged a large amount of total nitrogen and total phosphorus to the sea, resulting in severe eutrophication in the Changjiang River estuary and its adjacent sea areas, increasing the incidence of red tide (Tu et al., 2006). According to observations and statistics on global hypoxic zones by Diaz et al., eutrophication is the most critical factor contributing to the occurrence of anoxic zones in the sea (Diaz et al, 2001). Additionally, the Changjiang River’s diluted water, the Taiwan Warm Current’s spring layer, the organic matter delivered by the Kuroshio Current, and the presence of upwelling also exacerbate this situation (Zhou et al., 2010). The occurrence of an anoxic zone in the Changjiang River estuary in 2006 was due to hydrological conditions during the winter and spring that gradually matured in the summer, developing northward, and ultimately resulting in a north-south dual-core structure in July and August (Wei et al., 2015). For the environmental parameters obtained in this survey, the DO contents of all stations were greater than 2 mg l−1, and those in hypoxic zones were between 2 and 3 mg l−1; thus, this study purported to investigate the relationship between macrobenthic functional groups and environmental factors under low oxygen rather than hypoxic conditions.
Potential reasons for the difference of functional groups in different zones
Due to various factors such as diluted water and sediment deposition, the environmental factors of the Changjiang River estuary are much different from those of its adjacent sea areas. Extremely low salinity is the most profound environmental factor, the lowest level being 0.02, whereas the DO level is rather high, with a maximum level of 7.41 mg l−1 at some stations. High levels of nitrogen and phosphorus are also present. Moreover, environmental changes in the estuary area are more abrupt and more influenced by human factors. Therefore, those species that are more likely to inhabit the area are primarily eurythermal and euryhaline, e.g. Aglaophamus dibranchis, characterized by both low density and low biomass.
The years 2004 and 2006 were not typical of the trend of decreasing DO levels in the hypoxic zone (Figure 4) (Liu et al., 2012). DO levels in those years were affected by global El Nino events and typhoon during August (Chen et al., 2016), the disturbance in the water is violent, so DO is much higher than the past years. As a result, this study shows that persisitent hypoxic zones can recover their marine macrobenthic community in years when the environmental conditions prevent the hypoxia from occurring.
Levings demonstrated that globally, DO concentrations below 0.5 mg l−1 result in a high death toll among benthos in estuaries (Levings, 1980). Nilsson and Rosenberg et al. also found that after prolonged hypoxia, there were no signs of life in the sediments of coastal zones and bays (Nilsson and Rosenberg, 2000; Rosenberg et al., 2001). However, these findings were observed under long-term and extremely hypoxic conditions. Because the hypoxic region of the Changjiang River estuary changes seasonally, and the DO level is not lower than 2 mg l−1, the distributions of macrobenthic functional groups at different stations in the region exhibited no significant difference. Studies have shown that under hypoxic conditions, some polychaetes adapt to hypoxic environments by increasing their number and volume of gills to enlarge the respiratory surface area (Lamont and Gage, 2006). In addition, some organisms move from beneath the sediment to the surface layer to access more oxygen and mitigate the impacts of oxygen deficiency (Montagna and Ritter, 2006). Other studies indicate that after a region experiences hypoxia, the invasion of polychaetes will increase the abundance of benthos in the hypoxic recovery area (Karlsson et al., 2017). Interestingly, in a study on the macrobenthos response to hypoxia in northeastern Canadian waters, Rowan et al. found that the situation revealed by fossil materials is similar under prehistoric marine anoxic conditions (Rowan, 2017). The study of macrobenthic communities in the Changjiang River estuary’s anoxic zone revealed that the species community compositions of the non-hypoxic and hypoxic zones are not significantly different (Liao Y et al., 2017).
The data from this survey also showed that under slightly hypoxic conditions, organisms can cope with environmental changes through self-adaptation; thus, there were no significant differences in bio-density and biomass between hypoxic and non-hypoxic areas, with similar functional group compositions. However, from these data, we can still find some interesting phenomena. The bio-density of the detritivorous group in the hypoxic zone was higher than that in the non-hypoxic zone, but the biomass of the detritivorous group in the hypoxic zone was only one half that in the non-hypoxic zone, and this pattern was also present in the carnivorous group; thus, we speculate that hypoxic environmental conditions limited the growth of individuals within these two groups but not their ability to reproduce. In other words, among species in the same functional group, those with small body sizes but fast reproduction rates are more adaptive within hypoxic zones. Other studies show other studies show that the hypoxic zone may have been recently re-colonized after the zone become non-hypoxic by the detritivorous group, resulting in younger - smaller individuals (biomass divided by bio-density), or smaller species with a rapid reproduction and opportunistic life cycle (Yong Xu et al. 2016). Therefore, the herbivorous group, present in the non-hypoxic zone but absent in the hypoxic and estuary zones, reacted more profoundly, indicating that this functional group has higher environmental requirements, and slight environmental changes are likely to cause its demise. Conversely, the density and biomass of planktophagous benthos in the hypoxic zone were higher than those in non-hypoxic zone, indicating that they are more adaptive to live in the hypoxic zone. Eutrophication is an important cause of hypoxia; however, the red tide caused by high levels of nitrogen and phosphorus has led to a greater number of phytoplankton in seawater, providing a sufficient food source for planktophagous benthos.
Influence of environmental factors on functional group composition - Conclusions
The results of the CCA in this study showed that DO is not the most important environmental factor affecting the distribution of macrobenthic functional groups, whereas the contribution of DIN, temperature (T), or pH to the distribution is greater than that of DO (Fig. 3). In the Changjiang River estuary, the carnivorous group was the group most affected by DO, whereas the distributions of the detritivorous and omnivore groups were negatively correlated to DO content. The data for the stations in the non-hypoxic zone revealed that the distributions of the omnivorous and phytophagous groups were positively correlated to DO content and the detritivorous and carnivorous group distributions were highly positively correlated to temperature (T) and DIN. In the hypoxic zone, the distributions of the detritivorous and carnivorous groups were highly positively correlated to DO level, whereas the carnivorous group distribution was highly positively correlated to temperature (T), and the planktophagous and detritivorous groups were highly positively correlated to salinity (S). Multiple environmental factors simultaneously affected macrobenthos, leading to the aforementioned distribution patterns. The results from a similar study revealed that the macrobenthic distribution in the Changjiang River estuary and the Yellow and Bo Seas is a result of the interaction of various environmental factors, including human factors, such as organic pollution, ship navigation, etc (Jia et al., 2014).
This work was funded by the National Key Research and Development Program of China (2018YFD0900901), Scientific Research Fund of the Second Institute of Oceanography, SOA (JG1616), National Marine Public Welfare Research Project of China (201505004-3), National Natural Science Foundation of China (41706125), Natural Science Foundation of Zhejiang Province (LQ19D060004), Project of Long-term Observation and Research Plan in the Changjiang Estuary and Adjacent East China Sea (LORCE,14282).