To evaluate the ecological restoration effect of floating and submerged macrophytes on phytoplankton inhibition and water quality improvement, an enclosure experiment was undertaken in situ including Hydrilla verticillata mono enclosure, Eichhornia crassipes mono enclosure, and H. verticillata plus E. crassipes mixed enclosure. Concentrations of total nitrogen, ammonia nitrogen, nitrate nitrogen and total phosphorus were significantly lower in the vegetated enclosures than those in the control enclosure. Eichhorniacrassipes mono enclosure had significantly lower concentrations of ammonia nitrogen and nitrate nitrogen compared to the mixed enclosure. Purification ability was higher for faster growing macrophytes. The density and biomass of the phytoplankton in the plant enclosures were lower than those in the control enclosure, while the values in the mixed enclosures were significantly higher than those in the mono enclosures. Chlorophyta were the dominant algae in all the enclosures. Pyrrophyta and Cryptophyta were the co-dominant algae in the control enclosure. Therefore, macrophytes had a significantly inhibitory effect on phytoplankton, with the mono enclosures showing a stronger inhibitory effect compared to the mixed enclosures.
Many shallow lakes have become eutrophic as a result of anthropogenic influences. Eutrophication can decrease water transparency (Jeppesen et al., 1998; Scheffer, 1998), increase the occurrence of algal blooms (Glibert et al., 2001) and subsequently affect human health in the case of algal toxins (Davis et al., 2009; Hochmuth et al., 2014). It is urgent to alleviate and reverse eutrophication in degraded lakes, ponds and reservoirs with effective tools for most water resource managers. Macrophytes restoration has been widely applied in ecological engineering for the treatment of wastewater, owing to their efficacy in assimilating nutrients and creating conditions facilitating microbial decomposition of organic matter (Wang et al., 2009).
In ecological restoration, macrophytes, especially submerged macrophytes, are considered to be crucial because of multiple ecological functions such as regulating lake biological structure (Steffenhagen et al., 2011), improving the self-purification capacity (Blindow et al., 1993; Tanner and Headley, 2011; Soana et al., 2012), and stabilizing sediments and preventing resuspension (Scheffer et al., 1993). In current projects, floating macrophytes, floating-leaf macrophytes and submerged macrophytes are all selected. Although the restoration of submerged macrophytes is inevitably restricted by many factors such as water depth, wave action, water turbidity and transparency, and sediment physicochemical properties (Hamilton and Mitchell, 1996; Qin, 2009), many scholars have selected submerged macrophytes in ecological restoration since the 1980s (Wheeler and Center, 1996; Qiu et al., 2001). In contrast to some projects employing submerged macrophytes, floating macrophytes tended to be preferred in some projects (Song et al., 2011; Rahman and Hasegawa, 2011; Keizer-Vlek et al., 2014), especially in some urban small ponds. Floating plants were also adopted in some plant floating-beds systems as an important ecological remediation to control water eutrophication (Li et al., 2010).
However, comparative disadvantages in using submerged or floating macrophytes in lake restoration should be considered. First, it is difficult to control the growth area of floating plants in large field sites and second, floating plants are especially vulnerable to natural disasters in tropical and subtropical areas (Li et al., 2010). In contrast, it is difficult for submerged plants to survive in comparison with floating plants; they are more vulnerable to water conditions such as depth, and water transparency etc. (Li et al., 2013a).
Floating and submerged plants have been applied in different lakes, while integrated management projects are rarely formulated because there has been no mechanism to evaluate the relative benefits and drawbacks to the plants in relation to their perceived values (Johnstone, 1986), and there is little information on phytoplankton inhibition by the two kinds of macrophytes in a same system. In this study, we hypothesized that a mixed plant system with floating plants and submerged plants would promote the purification compared with a mono-plant system. The main objective of this study was: (i) to compare the effect between floating and submerged plants in water improvement and phytoplankton inhibition, (ii) to compare the effect between mono and mixed cultures in water improvement and phytoplankton inhibition.
Materials and methods
In situ field experiment system
An in situ field experiment was carried out in an ultra eutrophic pond located in Guangzhou, P.R. China, with a water depth of 1.5–2 m covering an area of about 2000 m2. Guangzhou has a subtropical climate, with an annual average air temperature of 23.2°C. The coldest month (January) and the hottest month (July) are 13.9°C and 28.7°C, respectively. The annual precipitation is about 1800 mm, with the wet season occurring from April to September and the dry season from October to March.
Twelve cylindrical enclosures, each with a diameter of 2.0 m and a height of 2.5 m, were constructed with a 0.2 cm high-density polyethylene film, and 8 PVC pipes within each enclosure for reinforcement. The base of the enclosures was inset 30 cm into the silt at the bottom of the pond, and the top was open and about 40 cm above the water level.
Four treatment groups were designed for 12 enclosures in May 2009. They included Hydrilla verticillata, Eichhornia crassipes, H. verticillata plus E. crassipes, and the control enclosure (without macrophytes) group. Each group had three repetitions. In E. crassipes enclosure (E enclosure), about 40% of the surface area was separated by PVC pipes for planting E. crassipes. In H. verticillata enclosure (H enclosure), H. verticillata was planted by 40% in the area of the sediment. In H. verticillata plus E. crassipes enclosure (H+E enclosure), H. verticillata and E. crassipes were planted 40%, respectively. Harvests were conducted in October 2009 and February 2010, about half of the standing biomass of both the submerged plants and the floating plants were harvested. The biomass was measured after they were dried at 80°C for 48 h.
Sampling and analysis
Water samples were taken from each enclosure monthly between 10:00–11:30 a.m. from July 2009 to June 2010 for water quality. A total of 144 samples were taken. Once a sample was collected, water temperature, pH, dissolved oxygen (DO) and transparency (SD) were measured immediately with a YSI MDS Multi Probe System (YSI Inc., USA). Unfiltered water samples were analyzed for TN, NH4-N, NO3-N and TP with an AA3 continuous flow analyzer (Bran+Lubee Inc., GER).
Phytoplankton samples for alga counting were obtained by taking 1000 mL of water from a 5 L pooled sample collected using a cylindrical sampler. The water samples were preserved with Lugol's iodine and sedimented for more than 48 h. The supernatant was removed and the residue of 30 mL was collected. The alga density in 0.1 mL concentrated samples from the mixed residue was counted directly through a 0.1 mL counting chamber using a microscope at 40 × 10 magnification. Alga biomass was estimated according to the closest geometric shape of each taxon. Identification of algae was performed to species or varieties according to Hu and Wei (2006).
All statistics were performed using Excel (Microsoft, Inc., USA). One-way ANOVA and the Least significant difference (LSD) multiple comparison test were performed to detect the significant differences under different treatments in nutrients concentration, phytoplankton density and biomass All results are presented as mean values ±SD. Differences were considered significant at P < 0.05.
Both H. verticillata and E. crassipes grew well after planting, and H. verticillata both in mono and mixed enclosures was able to extend to the surface by September 2009. However, the decay of dead plant material occurred with plant growth successively, the decay in the mixed enclosure occurred earlier, and the amount of rotten plant material was more than the mono enclosures. E. crassipes in the mixed enclosure began to rot in December 2009, while it began to rot 2 months later in the mono enclosures.
The control enclosure had the highest average TN concentration throughout the year, significantly higher (P < 0.05) than that in the plant enclosures (Table 1), that had no significant differences. TN concentration varied over the year, with the highest values in April and March 2010. TN concentrations of the three plant enclosures were highest in April 2010, and the value of the control enclosure was highest in November 2009 (Figure 1a).
Similar to the TN concentration, the highest concentration of TP also occured in the control enclosure, and it was also significantly higher than that in the plant enclosures (P < 0.05; Table 1), with no significant differences. The TP concentration in the plant enclosures varied significantly with vegetation growth, with higher values in the late part of the plant growth period (Figure 1b).
The macrophytes exhibited a strong NH4-N and NO3-N removal effect (Table 1). Both NH4-N and NO3-N concentrations in the control enclosure were significantly higher than those in the three plant enclosures (P < 0.05; Table 1). NH4-N and NO3-N concentrations in the mixed enclosure were significantly higher than those in the mono enclosures (P < 0.05; Table 1). Concentration of NH4-N in the control enclosure varied significantly with the seasons, with higher values in the winter months. As for the mono enclosures, concentration of NH4-N was higher before December 2009, while the mixed enclosure reached the highest value in February 2010 (Figure 1c). The concentration of NO3-N in the plant enclosures varied significantly with time, with higher values after April 2010 (Figure 1d).
Phytoplankton taxa and biomass
One hundred and twenty-one algae belonging to 7 taxonomic units were detected in 12 enclosures. Amongst 121 algae taxa, there were 66 taxa of Chlorophyta (54.5%), 23 taxa of Bacillariophyta (19.0%), 17 taxa of Cyanophyta (14.0%), and 15 taxa of another four different taxonomic units (12.5%). The control enclosure had significantly more algae taxa than the mono enclosures (P < 0.05; Table 2). The taxa of phytoplankton in the plant enclosures varied with macrophytes growth with more taxa during the vigorous growth period (Table 2). The dominant algae were Chlorophyta. Bacillariophyta and Euglenophyta were also major taxa in the three types of plant enclosures, while Cyanophyta, Pyrrophyta and Cryptophyta predominanted in the control enclosure.
The density of phytoplankton in the control enclosure was significantly higher than that in the plant enclosures, and the value in the mixed enclosure was also significantly higher than that in the mono enclosures (P < 0.05; Table 3). Density of the phytoplankton in the plant enclosures presented a marked seasonal variation, with lower values from July to September 2009 than the other months (Figure 2a).
As a whole, the biomass of the phytoplankton in the control enclosure was significantly higher than that in the three plant enclosures. The value in the E enclosure was significantly lower than that in other two plant enclosures throughout the experiment (P < 0.05; Table 3). The biomass of phytoplankton in the plant enclosures showed a rising trend with time, with the higher values occurring at the end of the experiment (Figure 2b).
Mono versus mixed culture
Our study confirms that the presence of macrophytes is of great importance for maintaining clear water in shallow waters, and our result has been reported by many previous studies (Arts, 2002; Dai et al., 2012); however, there are studies indicating something contradictory (Cardinal et al., 2014). Considering plant selection, some researchers supported species combination (Sirianuntapiboon and Jitvimolnimit, 2007; Li et al., 2010; Tanner and Headley, 2011; Zhang et al., 2012), while others prefer monocultures to avoid competition of different species with similar niche so that the overall effect would not be restrained (Doyle et al., 2003; Wu et al., 2007). Of the two species used in this study, the submerged H. verticillata can absorb nutrients from the sediment and water throught its root and leaf system, while E. crassipes can absorb dissolved nutrients from the water through its developed root system. So the combination effect would theoretically be better than a monoculture for the purpose of water purification. The actual result, however, was not as expected. The mixed enclosure exhibited a weaker purification effect than the mono enclosures.
Several possible explanations were considered. First, only two harvests were conducted throughout the whole experiment. Appropriate harvesting can decrease plant death, which is very important for protecting the lake water from secondary pollution and eutrophication (Xu et al., 2014). In this study, the concentration of TN and TP in the mixed enclosure was significantly lower than the two mono enclosures after the plants were harvested, so the lower frequency of harvest could be an explanation for why the mixed enclosure had a lower purification effect. A second explanation is that time of plant stable growth is limited, and even briefer in a mixed system, while plant stable growth in a long time is beneficial to the nutrients removal. In this study, the water depth of the enclosures was less than two meters and the surface area was small. Due to space limitation, macrophytes were easily corrupted after they reached their growth peak, and the decay and decomposition of plants would have had a deteriorating effect on water quality (Masifwa et al., 2001; Xu et al., 2014), and could have significantly influenced aquatic nutrient cycling, especially in eutrophic shallow waters (Li et al., 2013b). In this study, the earlier decay of plants and more rotten plant material would be the major reason for the mixed enclosure having higher nutrient concentrations than the mono enclosures.
Effect on phytoplankton
One of our main intentions was to control the algal blooms with the use of macrophytes. Additionally, attention was paid to reduce the concentration of chlorophyll-a in order to improve the water transparency. In this study, the density and biomass of the phytoplankton in the control enclosure were very high, several times those in the plant enclosures. The density and biomass of the phytoplankton in the E enclosure were the lowest among the enclosures, reflecting that phytoplankton was strongly inhibited in E enclosure. E. crassipes suppresses the phytoplankton growth through nutrient competition and shading (Zhou et al., 2014), allelochemical secretion (Zhao et al., 2012) and physical adsorption by a developed root system (Polprasert and Khatiwada, 1998; Kim and Kim, 2000). H. verticillata may compete with the phytoplankton for space and nutrients directly and provide a habitat for large zooplankton. In our study, floating macrophyte E. crassipes showed a stronger inhibition on phytoplankton than submerged macrophyte H. verticillata. Whether the same pattern displayed by other plants in the same system should be furtherly studied.
In this study, the effects on water quality improvement and phytoplankton inhibition of mixed enclosure were not better than the mono enclosures, which might be largely correlated with plant choice and special conditions such as water depth, plant harvest, etc. So for water restoration projects plant selection should be fully considered.
(1) Macrophytes can effectively restore eutrophic waters by significantly reducing the concentration of dissolved nutrients.
(2) Both the floating and submerged macrophytes species significantly reduced the density and biomass of phytoplankton. Phytoplankton in the vegetated enclosures was significantly reduced in comparison with the un-vegetated enclosures.
(3) The mono enclosures performed better than the mixed enclosure, and there were obvious differences between the two monocultures.
The authors would like to thank Zhong Tao, Liang Xiaodong, Wang Shuqiang and Zhang Chengfeng for their help during the experiment.
This work was supported by the National Science Foundation of China (31370476), the Fundamental Research Funds for the Central Universities (11613301), the Science and Technology Planning Project of Guangdong Province, China (2014A020217005), the project of the Urban Management Bureau of Shenzhen (201510) and the Baiyun District Water Supply Bureau Ecological Restoration Project for Madong Wetland Park Eutrophication Lake.