Three common plant species were collected from the Shuxi River in Taian for use in the experiment: Scirpus validus, Phragmites australis, and Acorus calamus. The test materials were collected in mid-April. Robust and healthy plants were collected, 50 individuals of each species; the sediment and plankton were removed from the surface of the roots with tap water in the laboratory, and the withered and yellow leaves were removed. Then, the plants were transferred to 20 L buckets with 15 L tap water; after exposure to the sun for five days, the plants underwent two weeks of cultivation and domestication in the artificial greenhouse of Shandong Agricultural University to become adapted to the greenhouse environment. After domestication, the plants were ready for the simulation test.

The culture solution was based on the Hoagland culture solution, with KNO₃ added; three nitrate concentrations were established―5 mg/L, 15 mg/L and 25 mg/L―and the 20 L blue plastic bucket was used as the culture container with the 15 L nutrient solution. Three individuals of the same species were fixed to the top with a foam board and wrapped in a black plastic bag to keep the bucket in a dark state. Each treatment included three repetitions; one plant was used as the control group, and Scirpus validus, Phragmites australis and Acorus calamus treatments were named S5, S15, S25; P5, P15, P25; and A5, A15, P25 in 5 mg/L, 15 mg/L and 25 mg/L nitrate concentrations, respectively. Nitrate was the sole nitrogen source, and denitrification or mineralization in water was the process of decomposition; the determination of the total nitrogen content of the water was used to indicate the removal of nitrate. At the start, a portable water quality analyzer (HI9828 HANNA, US) was used to determine the pH and DO; at the same time, water samples were taken and the total nitrogen content was determined with a TOC (N/C multi 3100, Germany) and the nitrogen content in plants was evaluated with an elemental analyzer (ECS4010 Costech, Italy). In all of the treatments, the experiment ended when the total nitrogen concentration in the water was lower than 2.0 mg/L, and the experiment lasted for 35 days. The nitrate removal percentage and plant purification capacity calculation formula is as follows:

% Nitrate Removal (W):

Purification ability of plants (M):

where C₀ is the initial concentration of nitrate (mg/L), C_i is the concentration of nitrate after 1 week under plant purification, and FW refers to the fresh weight of plants.

The data were entered into Microsoft Excel 2013 and an Origin 9.0 spreadsheet. All of the representative values were displayed as the mean value and standard deviation (S.D.). Significant differences among treatments were determined with an analysis of variance (ANOVA) in SAS 9.0. A P-value of less than 0.05 was considered statistically significant.

The three species of emergent plants with the same growth states were transplanted into different concentrations of nitrate, cultured for five weeks, and then measured for the fresh weight. The results indicate that all three species exhibited normal growth. All three species produced new roots, gemmae and adventitious roots during the test. At the end of the experiment, parts of the Phragmites australis leaves had withered and parts of the buds of Scirpus validus had also withered. However, Acorus calamus’ growth status was normal. Figure 1 shows fresh weight of the three species of macrophytes at the initial of test and the end of test. At the beginning of the experiment, the fresh weights of Scirpus validus, Phragmites australis and Acorus calamus were 38 g, 37 g and 101 g; after the test, the average weight gain was 21%, 8% and 26%, respectively. The weight gain of the emergent plants increased along with the increase in nitrate.

Table 1 shows that the removal efficiency of aquatic plants on nitrate is influenced by the concentration and plant species and the interaction between them (P = 0.000). Under different concentrations of nitrate, the removal efficiency of plants was significantly different (Table 2, P < 0.01); with increased concentrations of nitrate, the removal percentage of plants increased and the nitrate removal percentage was 81.62% at a concentration of

Note: different letters indicate significant difference (P<0.05) among the treatments.

25 mg/L. There was a highly significant difference among the three types of aquatic plants with respect to the removal percentage of nitrate (Table 2, P < 0.01). The nitrate removal percentage of Acorus calamus was highest, reaching 86.16%, and the nitrate removal percentage of Phragmites australis was significantly lower than the other two species, at only 62.34%.

Statistically significant increases in N removal were observed under increasing N concentrations in Scirpus validus in treated water. Among the concentrations, the highest mean nitrate removal occurred at 25 mg/L, with a total removal of 5.32 mg/L. At concentrations of 15 mg/L and 5 mg/L, the amounts removed were 4.44 mg/L and 2.14 mg/L, respectively. The highest (5.86 mg/L) mean N removal was recorded in the first weekly interval. However, removal during the 2nd, 3rd, 4th, and 5th weeks exhibited significant differences, other than the latter three weeks (Table 3). The weekly reduction of nitrate N by Scirpus validus under different levels of nitrate in water during the fourth week also exhibited a declining trend with increasing concentrations of nitrate in water. The highest percentage (79%) reduction was recorded at a concentration of 25 mg/L, with a removal amount of

Note: different letters indicate significant differences (P < 0.05) among the treatments.

19.68 mg/L. In comparison, 57% and 70% of N was removed from water under initial nitrate concentrations of 5 mg/L and 15 mg/L, respectively.

A comparison of the total nitrogen content of Scirpus validus obtained from different N concentrations in water indicate that plants growing under higher concentrations had a higher N content in their tissues (Table 3). The total N content of Scirpus validus was 22.27 mg/g, 21.77 mg/g, and 21.03 mg/g in water with nitrate concentrations of 25 mg/L, 15 mg/L and 5 mg/L, respectively, and the difference was not statistically significant.

There was a major increase in nitrate removal with increasing nitrate concentration in water. On average, 2.54 mg/L, 6.12 mg/L and 5.14 mg/L were removed under nitrate concentrations of 5 mg/L, 15 mg/L, and 25 mg/L, respectively. The results exhibited a decreasing trend in N mean removal per week over time, apart from the fourth week. A higher (6.39 mg/L) mean nitrate removal percentage was observed in the first week, and the amount removed was 2.56 mg/L and 0.5 mg/L in the second and third week, respectively. The percentage of nitrate removal by Phragmites australis was also studied under different water N concentrations. An obvious increase in N removal was observed with increasing water N concentrations. At the end of third week, 2.54 mg/L, 6.12 mg/L and 5.14 mg/L N were removed from the water under initial N concentrations of 5 mg/L, 15 mg/L and 25 mg/L, respectively (Table 4). The total N content of Phragmites australis was observed at 21.67 mg/g, 20.40 mg/g, and 20.33 mg/g in tissues grown in water with N concentrations of 25 mg/L, 15 mg/L and 5 mg/L, respectively. The highest (21.67 mg/g) total N content was recorded in plants grown in water with a nitrate concentration of 25 mg/L, but the difference was not statistically significant.

Table 5 shows the mean values of nitrate removal for the different levels of nitrate for a given week. The average mean N removals of 3.39 mg/L, 12.59 mg/L and 21.68 mg/L were recorded in water with initial concentrations of 5 mg/L, 15 mg/L and 25 mg/L, respectively. In the case of Acorus calamus, the highest (8.13 mg/L) amount of nitrate was removed in the initial week, followed by the fifth week (2.15 mg/L) and the second week (1.29 mg/L). The same removal pattern was exhibited by the other two species, for which the highest nitrate removal was recorded during the initial week. However, even in the final week, substantial removal was achieved by Acorus calamus; this could be attributed to the high N requirements for developing the root system in this species. Acorus calamus exhibited a general increasing trend of nitrate removal from water with increasing water nitrate concentrations. The total N content of Acorus calamus was observed at 21.67 mg/g, 20.40 mg/g, and 20.33 mg/g in tissues grown in water with N concentrations of 25 mg/L, 15 mg/L and 5 mg/L, respectively. The highest percentage (87%) reduction occurred under a concentration of 25 mg/L, whereas the lowest percentage (68%) reduction was recorded under a concentration of 5 mg/L.

The purification ability of different species under the same concentration of nitrate and the purification ability of the same species under different concentrations of nitrate are illustrated in Figure 2. Under the same concentration of nitrate, the three species exhibited significant differences in nitrate purification capacity (except P25 and S25). The purification ability of Acorus calamus was substantially lower than those of the other two species under an initial nitrate concentration of 15 mg/L, which was 49.60% for Scirpus validus and 43.28% for Phragmites australis, whereas these values were 39.75% and 41.29%, respectively, under a concentration of 25 mg/L. The purification ability of the same species under different nitrate concentrations were significantly different (P < 0.05). With an increase in the concentration of nitrate, the purification capacity increased, and the purification

abilities of Acorus calamus, Scirpus validus, Phragmites australis were 4.85-fold, 6.75-fold, 7.93-fold higher under nitrate concentrations of 25 mg/L than under nitrate concentrations of 5 mg/L, respectively.

At the beginning of the experiment, the pH of the water was 6.8, which changed over time; the pH in the treatment of different nitrate exhibited a trend of first increasing and then decreasing, finally returning to close to the initial pH (Table 6). Early in the experiment, the water pH increased dramatically. In the second week, the pH appeared to be at a maximum; in the Acorus calamus system, the pH was lower than that in the reed and bulrush systems: the pH of the reed and bulrush system was between 9.22 and 9.77, whereas that of the Acorus calamus system was between 8.56 and 8.71, both of which were significantly different from the pH in the control treatment. After two weeks, all of the treatments decreased in pH. The pH decreased to the lowest value in the fourth week, but a slight increase occurred in the 5th week. Overall, at the end of the experiment, the pH was higher than the initial level, and the pH was significantly higher than in the control group.

Table 6 shows that DO peaked in the water at two weeks and four weeks. At the beginning of the experiment, the DO in each treatment increased, reaching a maximum value at two weeks, and the DO content in each treatment was higher than in the control. The DO content was highest in the reed system, at 12.18 mg/L DO, which gradually decreased with time. Among the treatments without significant changes, at two weeks before the test, the DO content in the water was significantly lower than in the control, but three weeks later, the DO content began to decrease. At the end of the experiment, the content of DO in all three plant systems was lower than in the control, indicating that none of the three aquatic species resulted in a notable increase in oxygen containing nitrate in water.