Ecological Efficiency of the Mussel Hyriopsis cumingii (Lea, 1852) on Particulate Organic Matter Filtering, Algal Controlling and Water Quality Regulation

Abstract

Hyriopsis cumingii plays an important role in aquatic ecosystems and fishery economy due to its potential value for water purification and edibleness. The present study was undertaken to reveal the short- and long-term purifying effects of H. cumingii on pond water. The short-term experiment results showed that total suspended solids, particulate organic carbon, and fatty acid in water were significantly eliminated by H. cumingii, with average percentage of filtering effect respectively reaching 22.19%, 57.48%, and 21.00% in the high-density group. Analogously, H. cumingii could significantly reduce phytoplankton biomass, species number, density, and chlorophyll-a concentration, especially in the control of diatoms, green algae, and cyanobacteria. Besides, electric conductivity also was significantly reduced by H. cumingii and its variation tendency showed a typical density-dependent effect. Similar purification effects were observed in chemical oxygen demand (COD), total nitrogen, and total phosphorus. In the long-term monitoring test, H. cumingii has also shown positive effects on the water transparency and the control to phytoplankton and COD in recirculating aquaculture pond. Overall, H. cumingii showed excellent ecological function, and the mixed breeding density of shellfish in the recirculating aquaculture pond is worth further exploration.

1. Introduction

Human beings have complex social, economic, and ecological dependence on freshwater ecosystems, but the development of intensive aquaculture has brought potential threats to freshwater environments [1,2,3].

Freshwater shellfish, as a species able to purify water, has been neglected for a long time in China, with the production of 0.20 million tons only accounting for 0.66% of total freshwater aquaculture production in 2018 due to its small scale of breeding and long economic cycle [4]. Recently, resulting from the guidance of new environmental protection policies [5], the ecological function of freshwater shellfish has gradually attracted the attention of the aquaculture industry in China. Therefore, different aquaculture modes such as “benthic or hanging aquaculture of shellfish” and “macrophytes and bivalve mixed breeding” have been mentioned and developed again for freshwater purification [6,7].

Biomanipulation is an important way to restore eutrophic water bodies in the long term and to reduce costs can continuously regulate nutrient loading through different biological methods [8]. Among them, filtration by bivalves can lead to a large decrease in phytoplankton and other particles in the water column [9]. However, the filtration rate in freshwater or seawater is varied with different shellfish species and size, salinity, temperature, particle size and concentration, density, and flow regime [10,11,12]. Furthermore, shellfish or other mollusks may control microbial concentrations and community structure either directly by using bacteria as a food source or indirectly by producing fecal or pseudo-fecal material that affects bacterial activity and growth [13,14]. Nevertheless, research about freshwater shellfish is still in its infancy and its ability to filter aquaculture water and related water purification mechanisms need to be studied urgently.

The bivalve mollusk triangle sail mussel (Hyriopsis cumingii), which belongs to the Unionidae family, is widely used for pearl production. At present, despite obvious progress in the research of genetic breeding [15], growth traits, and pearl cultivation [16], the application of H. cumingii for water purification are still left behind, although the culture of excessive filter-feeding bivalves may harm the balance of the ecosystem [17,18]. In recent years, the use of H. cumingii has been proposed as an effective strategy to control algal bloom in eutrophic water bodies, and several regulatory effects have been achieved [19]. The filter-feeding activity of H. cumingii may potentially support submerged macrophyte growth by reducing cyanobacterial density, but bivalve communities in natural water are in decline as a result of overharvesting [20]. Moreover, H. cumingii can also be used as a superior material for retrospective monitoring of eutrophication-induced environmental changes in aquatic ecosystems [21].

Overall, aquaculture water pollution is becoming a severe environmental issue throughout the world. The utility of bivalves as a bio-manipulation way to improve water quality and control algal blooms has been broadly acknowledged, but the potential purification mechanism of bivalve shells has received little attention. This study aimed to evaluate the short- and long-term impacts of H. cumingii on pond water. The purpose of this study was to promote the development of the ecological aquaculture model based on H. cumingii in suitable inland areas of China.

2. Materials and Methods

2.1. Shellfish and Experimental Design

In a short-term experiment, healthy H. cumingii with similar characteristics (wet weight: 74.7 ± 6.0 g, shell length: 8.2 ± 1.0 cm) were obtained from the Comprehensive Experimental Station of Jinhua, National Shellfish Industry Technology System (Zhejiang, China) and were maintained in an aquaculture pond, with temporary breeding for a month at natural temperature. Five groups with different densities of 0 individuals (CG), 3 individuals (G1), 6 individuals (G2), 9 individuals (G3), and 12 individuals (G4) were set for the short-term test of filtering water. Each group was performed in triplicate. The experiment was conducted in continuously oxygenated opaque plastic drums filled with 100 L of pond water that came from an aquaculture pond of about 15,000 m² on campus. The short-term experiment was carried out for a week continuously in a closed-form without fish being cultured.

In the long-term experiment of monitoring aquaculture pond, H. cumingii individuals were placed behind the tailwater discharge area of the recirculating aquaculture pond (total 5 flume, the total area is about 2.93 hm²) in Xinyue Aquaculture Co., LTD, Bishan District, Chongqing, China (Figure 1). Barbus capito is the main fish species cultured in the flume. The average diet was about 2% of the fish’s body weight. In each flume (length: 7 m, width: 2.5 m, water depth: 2 m), it was set about 500 individuals of (average wet weight: 368.9 ± 106.1 g, shell length: 15.7 ± 1.4 cm) H. cumingii in the purification area. Collecting of phytoplankton and water samples of the tailwater discharge area, H. cumingii purification area, and ecological purification area were conducted monthly except in February and November in 2018 (15 January, 21 March, 18 April, 28 May, 20 June, 19 July, 12 August, 4 September, 11 October, 25 December), as considering with the water quality condition is stable in low-temperature season. Two independent samples were acquired and detected at a time per sampling site.

2.2. Particulate Organic Matter Detection

In the short-term experiment, total suspended solids (TSS) were measured by the modulated gravimetric method [22]. Briefly, 100 mL of the water sample from each treatment was collected and adequately filtered by 0.22 μm membrane filtration, then immigrated into 103–105 °C of drying oven for an hour. After that, the samples were cooled and weighed. Particulate organic carbon (POC) was measured by the combustion oxidation and non-dispersive infrared analysis method [23] using TOC Multi N/C 2000 analyzer (Analytik Jena, Germany). Fatty acid (FA) was determined by LC-20A HPLC System (Shimadzu, Tokyo, Japan). A more detailed description is given as follows: each 1 L water sample was sufficiently extracted using the chloroform-methanol mixture and concentrated using rotary evaporators. Then samples were collected and dissolved in mobile phase solution (V_Methanol: V_{Monopotassium phosphate solution (0.02 M)} = 5.95) for fatty acid determination (SPD-20A UV detector, Shimadzu, Tokyo, Japan) at 210 nm, with three replicates. The percentage of filtering effect on particulate organic matter (TSS, POC and FA) was calculated according to the following equation:

$$F=\left(1-\frac{P_i}{P_C}\right)\times 100\%$$

(1)

where F is the percentage of filtering effect; P_i is the particulate organic matter concentration in the treatment group; P_c is the concentration in the control group.

2.3. Chlorophyll-a and Phytoplankton Quantitative Detection

To explore the effects of H. cumingii on chlorophyll-a (Chl-a), each 0.5 L water sample of control and H. cumingii groups were collected in the short-term experiment, then filtered with 0.45 μm filter membrane and extracted with 90% acetone, centrifuged at 3500 rmp and each supernatant was removed. All samples were measured following acetone spectrophotometry (Khaleghi, 2020). The calculation formula is shown as follows:

Chl-a = (11.85 × (A₆₆₄ − A₇₅₀) − 2.16 × (A₆₄₅ − A₇₅₀) + 0.10 × (A₆₃₀ − A₇₅₀)) × V₁ × V × δ

(2)

where A is the absorbance at different wavelengths; δ is the optical spectroscopy path length of cuvette; V₁ and V are the water sample volume and extraction volume, respectively.

The phytoplankton examination was performed in the short- and long-term experiments. The detailed description is given as follows: 1 L of water samples was collected and preserved immediately with Lugol’s iodine solution, precipitated for 24 h, and then concentrated to 50 mL volume. The phytoplankton qualitative and quantitative examination was assessed by observing and counting with a microscope (Leica DM500, Wetzlar, Germany). The identification of phytoplankton genus and species was conducted by referring to “The common phytoplankton atlas of Chongqing region of the three gorges reservoir area” and “The common freshwater phytoplankton algae atlas of China” [24,25]. Phytoplankton density, biomass, and statistical calculations were carried out according to pieces of literature [26,27] and the quantitative formula of phytoplankton is given as follows:

$$C= N\times \frac{S_c}{S_f\times n}\times \frac{V_1}{V_2\times V}$$

(3)

where C is the phytoplankton density; N is the number of phytoplankton in the observation field; S_c is the area of the haemocytometer; S_f is the area per view; n is the number of views per slice; V₁ is the volume after concentration; V₂ is the volume of the counting box and V is the volume before concentration.

Phytoplankton biomass (PB) was calculated by methods of cell volume conversion based on its size and average density [28]. Moreover, the species diversity and evenness were analyzed by using the Shannon-Wiener index (H). The biological diversity was calculated according to the following equation [29].

$$\mathrm{H}\prime =-{\displaystyle \sum }_{i=1}^n\left(n_i/\mathrm{N}\right)\ln \left(n_i/\mathrm{N}\right)$$

(4)

where n_i is the individual number of alga i, N is the total number of all species at each H. cumingii group.

2.4. Hydrochemical Parameters Detection of Samples

Water samples were collected and brought back to the laboratory for analysis of water quality according to relevant national standards [30]. In the short-term experiment, electric conductivity (EC) was monitored by a portable dissolved oxygen meter (HQ30D, HACH Company, USA). In the long-term experiment, water temperature and transparency were assessed using a water thermometer and Secchi disc, respectively. Moreover, total nitrogen (TN), total phosphorus (TP), and chemical oxygen demand (COD) were used for the quantitative evaluation of the trophic state in the pond water. Among them, TP was measured at 700 nm by the ammonium molybdate spectrophotometric method [31] with an ultraviolet spectrophotometer (UV-1800, Shimadzu, Japan). TN was measured at 220 nm and 275 nm by the alkaline potassium persulfate method [32] with an ultraviolet spectrophotometer (UV-1800, Shimadzu, Japan). Meanwhile, COD was measured by the dichromate method [33], and the calculation equation is shown as follows:

$$\mathrm{COD}({\mathrm{O}}_2,\mathrm{mg} {\mathrm{L}}^{-1})=\left[\left(V_0-V_1\right)\times \mathrm{c}\times 8 \mathrm{g}/\mathrm{mol}\times 1000\right]/V$$

(5)

where c is the standard solution concentration of ammonium ferrous sulfate; V₀ is the initial volume of ammonium ferrous sulfate standard solution; V₁ is the water sample titration volume of ammonium ferrous sulfate standard solution; V represents water sample volume.

2.5. Statistical Analysis

Statistical analyses in particulate organic matters, Chl-a, and water quality factors were performed using SPSS 16.0 software (SPSS Inc., Chicago, IL, USA). Data over the three replicate drums in the short-term experiment were expressed as the arithmetic mean ± standard deviation (SD) and analyzed by one-way ANOVA. The paired differences between treatments were determined by LSD-test, where p values less than 0.05 were accepted as a significant difference and expressed with a different letter in tables and figures. Each data in the long-term experiment was represented the average value basing on the same season, including Spring (21 March, 18 April, 28 May), Summer (20 June, 19 July, 12 August), Autumn (4 September, 11 October), and Winter (25 December, 15 January). The paired results of tailwater discharge area, H. cumingii purification area, and ecological purification area were grouped for preliminary comparison of water purification effect in four seasons.

3. Results

3.1. Short-Term Effects of H. cumingii on TSS, FA, and POC

Purification effects of H. cumingii on the particulate organic matter are presented in

Table 1. Total suspended solids (TSS), fatty acid (FA), and particulate organic carbon (POC) in different mussel density treatments.

	CG	G1	G2	G3	G4
TSS (mg L−1)	50.33 ± 6.08 a	46.33 ± 7.81 a	40.67 ± 2.52 a,b	39.00 ± 10.26 a,b	30.67 ± 3.79 b
FA (mg L−1)	16.6 ± 0.04 a	17.31 ± 0.02 b	12.38 ± 0.25 c	12.22 ± 0.19 c	10.48 ± 0.04 d
POC (mg L−1)	10.72 ± 0.03 a	4.80 ± 0.13 b	4.29 ± 0.12 d	4.41 ± 0.11 c	4.73 ± 0.23 b

Note: CG, G1, G2, G3 and G4 represent 0 mussels, 3 mussels, 6 mussels, 9 mussels and 6 mussels, respectively. Different lowercase letters denote significant difference between different treatments (p < 0.05).

. The results show that TSS, FA, and POC in the control group reached 50.33 ± 6.08 mg L⁻¹, 16.60 ± 0.04 mg L⁻¹, and 10.72 ± 0.03 mg L⁻¹, respectively. The results also reveal that H. cumingii individuals could significantly reduce TSS, FA, and POC (p < 0.05). Among them, 12 individuals (G4) of H. cumingii could significantly reduce TSS compared to the treatment without mussel (CG), with its percentage of filtering effect reaching 39.06%. Although density-dependent reductions were observed in the G1, G2, and G3 groups, they were not significantly different from the control group (p > 0.05). Although FA concentration in G1 significantly higher than that in CG, H. cumingii significantly reduced that in G2, G3, and G4 groups, with the highest filtration effect appeared in G4 (p < 0.05). Moreover, compared to CG, the concentration of POC was significantly dropped (p < 0.05) in groups in all treatments with mussel, and the filtering effect of H. cumingii was from 55.2% to 60.0%.

Overall, H. cumingii showed excellent filtration effects on TSS, FA, and POC in pond water, and its effects were associated with its density.

3.2. Filtering Effect of H. cumingii on Phytoplankton

Short-term effects of H. cumingii on Chl-a and PB were represented in Figure 2. Among them, the concentration of Chl-a in CG reached 9.97 μg L⁻¹. The concentration of Chl-a in treatment groups from G1 to G4 was significantly lower than that of the control group labeled as CG. However, there was no significant difference between the treatment groups, with a percentage of filtering effect from 45.2% to 61.1% (Figure 2A).

A total of 29 species of planktonic algae belonging to Cyanophyta (4 species), Chlorophyta (14 species), Bacillariophyta (10 species), and Chrysophyta (1 species) were found in this study. However, Mallomonas candata Iwan belonging to the Chrysophyta was only found in G4 (Figure 2B). In detail, the PB in CG reached 1.04 mg L⁻¹ and the proportions of Bacillariophyta, Chlorophyta, and Cyanophyta in the control group were 89.3%, 4.4%, and 6.3%, respectively.

The short-term experiment results indicated that H. cumingii individuals could effectively reduce the total PB in all treatment groups with mussel, and this downward trend of the total PB was consistent with the detection results of Chl-a. The total biomass of Bacillariophyta, Chlorophyta, and Cyanophyta was decreased to different degrees in all treatment groups. Moreover, phytoplankton community characteristics, including species number, density, and diversity index were shown in

Table 2. Phytoplankton community characteristics in different treatments.

Phytoplankton Statistical Analysis	CG	G1	G2	G3	G4
Bacillariophyta Biomass (10−3 mg L−1)	922.20	115.66	55.04	15.15	39.30
Navicula simplex Krassk	386.43	17.97	10.48	5.99	1.50
Synedra acus Kützing	239.65	31.95	17.97	5.99	17.97
Synedra ulna Ehrenberg	103.85	10.65			3.99
Diatoma elongatum Ag.	83.88	2.66	4.99	1.00	3.00
Melosira granulata Ralfs	67.10	51.12	20.77		10.38
Navicula cincta Van Heurck	27.26	1.30	0.65		1.95
Cocconeis placentula Hust	4.99
Cyclotella meneghiniana Kütz	3.99			2.00
Stauroneis kriegeri Patrick	3.06		0.17	0.17	0.51
Gomphonema constrictum Grunow	2.00
Chlorophyta Biomass (10−3 mg L−1)	45.63	30.49	10.28	4.94	36.43
Gonium sociale	27.56	14.38	5.99	3.00	27.56
Planktosphaeria gelatinosa Smith	4.49	6.66	1.65
Crucigenia puadrata Morren	3.99	3.20	1.20
Sceaedesmus cavinaus Chod	3.20	2.66	0.85		0.80
Crucigenia tetrapedia Morr	2.40	1.07	0.40	1.20
Schroederia spiralis Korsch	2.00
Selenastrum gracile	1.20	1.33			0.60
Ankistrodesmus angustus B.S. Korš	0.40	0.27	0.20	0.20	0.20
Cosmarium subtumidum Nordst	0.40
Sceaedesmus oblipuus Kütz					0.80
Sceaedesmus dimorphus Kütz		0.93		0.55
Pediastrum tetras Ralfs					1.60
Pediastrum duplex Mey					1.60
Cyanophyta Biomass (10−3 mg L−1)	64.44	38.30	17.77	0.53	1.64
Nostoc communes Vauch	43.94	13.31	15.98
Chroococcus minor Näg	16.11	3.15	0.59	0.15	1.63
Phormidium tenue Gom	4.39	21.83	1.20
Merismopedia punctata Meyen				0.38	0.01
Species number	22	17	15	11	16
Total Biomass (10−3 mg L−1)	1032.27	184.45	83.09	20.62	77.09
Phytoplankton Density	12.00	5.53	3.06	1.72	2.01
Diversity index H’	3.61	3.23	3.35	1.49	2.85

. In the control group, the biomass of nine kinds of plankton algae was more than 0.01 mg L⁻¹, including Navicula simplex Krassk (37.4%), Synedra acus Kützing (23.2%), Synedra ulna Ehrenberg (10.1%), Diatoma elongatum Ag (8.1%), Melosira granulata Ralfs (6.5%), Navicula cincta Van Heurck (2.6%), Gonium sociale (2.7%), Nostoc communes Vauch (4.3%) and Chroococcus minor Näg (1.6%). Moreover, the number of observed phytoplankton species in all treatment groups was 22 (CG), 17(G1), 15(G2) 11(G3) and 16(G4). The density of phytoplankton was dropped sharply from 12 × 10⁵ cell L⁻¹ to a range from 5.53 to 1.72, respectively, with its alteration trend being consistent with the detection results of Chl-a and PB. For the diversity of phytoplankton, the Shannon-Wiener index in G3 and G4 groups were lower than that of CG, G1, and G2 groups. In the long-term experiment, PB was greater than 25 mg L⁻¹ in spring, Summer and Autumn.

In general, the number of species, biomass, and diversity index of phytoplankton could be reduced effectively by the filtering effects of H. cumingii.

3.3. Effect of H. cumingii on the Water Conditions

Short-term effects of H. cumingii individuals on the EC, COD, TN, and TP were given in Figure 3 and Figure 4.

In detail, the EC showed a significant reduction (p < 0.05) in all treatment groups from G1 to G4 and its variation tendency showed a typical density-dependent effect in the whole experimental period (Figure 3A). Our results also revealed that at least six individuals (G2) of H. cumingii could significantly reduce (p < 0.05) the COD values compared to the treatment without mussel (CG) (Figure 3B) and a similar effect was found in TN content (Figure 4A). For TP content, however, it only showed a significant decrease (p < 0.05) in the treatments with at least 12 individuals (G4) of H. cumingii (Figure 4B).

The results indicated that H. cumingii individuals were able to reduce the EC, COD, TN, and TP of water.

3.4. Annual Survey Results in the Recirculating Aquaculture Pond

In the long-term experiment, the annual variations in temperature and pH of recirculating aquaculture water were shown in Figure 5A. Among them, the average water temperature in autumn reached 29.7 °C, and in winter, that was as low as 14.2 °C. Moreover, it in all seasons exceeded 20 °C except for winter. The water quality of the pond was weakly alkaline throughout the year. Thus, the results showed that H. cumingii individuals adapted to the year-round changes of pH and temperature in the recirculating aquaculture pond.

The purifying effects of H. cumingii individuals on the PB, transparency, COD, TN, and TP were reflected in Figure 5B–F. The same as the variation trend of water temperature, PB remained above 20 mg L⁻¹ from spring to autumn. H. cumingii individuals showed a potential filtering effect on phytoplankton based on visual interpretation of Figure 5B, as the PB in the ecological purification area was lower than that in the tailwater discharge area. On the other hand, due to their filtering effects on phytoplankton and particulate organic matter, the transparency of the ecological purification area was higher than that of other areas based on visual interpretation of Figure 5C. Identical to the PB, the COD value of the ecological purification area in each quarter was reduced compared with that in other areas, which was consistent with the short-term experiment results (Figure 5D). However, different from the short-term experiment, H. cumingii individuals did not reduce the TN and TP concentration, but increased in autumn and winter (Figure 5E,F).

4. Discussion

Since the beginning of artificial propagation in the middle of the 1980s, H. cumingii has become a dominant species of cultivating pearl in freshwater and is mainly distributed in China’s mid-east regions. However, only a few studies are involved in proper growth and pearl cultivation of H. cumingii [34]. Our newly approved Chinese Shellfish System Chongqing Comprehensive Experimental Station (2018), as the unique freshwater comprehensive experimental station in the non-main producing areas, is mainly responsible for exploring the ecological service function and fishery economic value of freshwater shellfish. Moreover, this study can further develop the shellfish industry based on the recirculating aquaculture model.

In the present study, H. cumingii individuals could effectively reduce TSS, POC, and FA in the aquaculture pond, with average percentage of filtering effect respectively reaching 22.19%, 57.48%, and 21.00% in the high-density group. The effects of H. cumingii was typically density-dependent on TSS, FA, and POC. The results indicated that H. cumingii profoundly regulated the substance cycle by direct filtration for suspended particulate organic matter in water.

Autochthonous particulate organic matter sources, such as phytoplankton, contained a higher percentage of labile organic composition than allochthonous sources, making it be as an appropriate proper source for primary consumers [35,36]. The organic carbon is also derived from allochthonous and autochthonous sources, including particulate and dissolved forms, and irradiation could realize the transformation of dissolved organic carbon to organic carbon particles [37,38]. Moreover, fatty acid directly or indirectly from primary producers, such as phytoplankton and photosynthetic bacteria, was an important evaluation index of primary water productivity [39]. However, the superfluous particulate organic matter could result in the deterioration of water quality and water eutrophication in the exoteric river estuary and enclosed pond water system [40]. Therefore, the effective reduction in particulate organic matter may be attributed to the direct filtration of H. cumingii for phytoplankton and organic debris.

Further research indicated that H. cumingii could effectively control the biomass of diatoms, green algae, and cyanobacteria, the primary food sources of bivalve shellfish. Especially in the control of the nine kinds of plankton algae with their concentration being more than 0.01 mg L⁻¹, they were Navicula simplex Krassk, Synedra acus Kützing, Synedra ulna Ehrenberg, Diatoma elongatum Ag, Melosira granulata Ralfs, Navicula cincta Van Heurck, Gonium sociale, Nostoc communes Vauch and Chroococcus minor Näg. Among them, six species of plankton diatoms made up 87.9% of PB in the pond water. Furthermore, H. cumingii filtered 87.3 percent of the six species of plankton diatoms. Briefly, H. cumingii showed a powerful filtering function on phytoplankton as its resulting filtration rates were greater than 82% in all treatment groups from G1 to G4, and this downward trend was consistent with the detected values of Chl-a. Therefore, the use of H. cumingii has been proposed as an alternative way to control algal bloom in eutrophic water in recent years [19].

As an essential indicator of the PB, Chl-a has been often used to evaluate water quality [41]. Studies had revealed that the filter-feeding H. cumingii may indirectly promote submerged macrophyte growth by controlling the density of algal bloom [20]. However, this study confirmed that H. cumingii considered as the primary consumer could effectively filter phytoplankton and was of great significance to nutrient utilization and the control of algae blooms in aquaculture.

Annual monitoring results show that transparency in the ecological purification area was higher than that in the tailwater discharge area due to the effect of H. cumingii on phytoplankton and particulate organic matter in the recirculating aquaculture pond. The other results show that co-cultivation of an aquatic plant (Ipomoea aquatica) and aquatic animal (H. cumingii) could effectively enhance the removal rates of the pollutant from the ecological floating bed contained with artificial medium [42]. These results also pointed out that the filter-feeding H. cumingii promoted the solubilization and mineralization of particulate organic matters, and enhanced the purification efficiency from the ecological floating bed by purifying microorganisms [42]. In eutrophic water, phytoplankton using N and P as nutrient sources generally grows rapidly and easily, resulting in algal bloom [43]. The concentration of TP from treatment groups in the short-term experiment showed a decreasing trend with the increase in mussel density. However, in the long-term experiment, H. cumingii individuals did not reduce the TN and TP concentration, but increased in autumn and winter. This result may be caused by climate change and weakened body functions. In general, this study revealed the potential role of H. cumingii in water purification, including increasing the transparency of water and reducing the EC and COD.

5. Conclusions

The overall water purification mechanism of H. cumingii remains uncertain, but current results have indicated that H. cumingii is an excellent way for the filtration of particulate organic matter, the control of algal bloom, and the regulation of water quality in aquaculture. Results from the short-term experiment and long-term monitoring experiment indicated that the model of culturing H. cumingii based on recirculating aquaculture system is of great significance in the ecological purification and healthy aquaculture. As potential sink of CO₂, H. cumingii furtherly may contribute to the reduction in greenhouse gases. In summary, this study provides a feasible basis for us to carry out the ecological cultivation and application of H. cumingii in the southwest of China in the future.