Effective dilution rate to suppress the risk of Microcystis blooms in Lake Tega, Japan, based on a competitive growth simulation model Open Access

N

Nitrate-nitrogen concentration (mg-N L⁻¹)

N₀

Initial nitrate-nitrogen concentration (mg-N L⁻¹)

P

Phosphate-phosphorus concentration (mg-P L⁻¹)

P₀

Initial phosphate-phosphorus concentration (mg-P L⁻¹)

C

Cell density (cells mL⁻¹)

Q

Cell quota (pg cell⁻¹)

ρ_max

Highest uptake rate(pg cell⁻¹ day⁻¹)

ρ_max^hi

Highest uptake rate at the beginning of cultivation (pg cell⁻¹ day⁻¹)

ρ_max^lo

Highest uptake rate at the maximum growth rate (pg cell⁻¹ day⁻¹)

K_μ

Half-saturation constant for growth rate (mg L⁻¹)

K_ρ

Half-saturation constant for uptake rate (mg L⁻¹)

K_ρ^hi

Half-saturation constant for uptake rate at the beginning of cultivation (mg L⁻¹)

K_ρ^lo

Half-saturation constant for uptake rate at the maximum growth rate (mg L⁻¹)

μ_max

Maximum growth rate at a certain nutrient concentration (day⁻¹)

μ′_max

Highest specific growth rate (day⁻¹)

Q_min

Minimum cell quota (pg cell⁻¹)

Q_max

Maximum cell quota (pg cell⁻¹)

D

Dilution rate (day ⁻¹)

d

Daily renewal rate (day⁻¹)

α

Inter-specific competition rate

Subscripts A and B, respectively, refer to two cocultured algal species (Microcystis sp. and Cyclotella meneghiniana)

Subscripts n and p, respectively, refer to nitrate-nitrogen and phosphate-phosphorus

The severe impact of cyanobacterial blooms in eutrophic lakes on water quality has caused a lot of environmental problems worldwide in recent decades. Many treatment methods to remove cyanobacterial blooms, such as pump suction, bubble floatation, chemical coagulation and so on, have been carried out and examined in several studies (Bui et al. 2015; Cottingham et al. 2015; Türkmen & Kütük 2017; Farnham et al. 2019). Water transfer is one of the effective treatment methods among them, not only for suppressing the formation of cyanobacterial blooms but also for ameliorating the water quality. The transferred water can wash the algae away and dilute or enrich the nutrient concentrations of water to inhibit or promote algal growth (Amano et al. 2012; Sugimoto et al. 2016). Although there have been many studies focusing on the inhibitory effect of water transfer on cyanobacterial blooms, the limited experience in some certain lakes could not be applied to other lakes worldwide universally.

From 1974 to 2000, the lake ranked with the worst water quality in Japan was Lake Tega (35° 51′ N, 140° 02′ E) (Chiba Prefectural Government 2019), which suffered from cyanobacterial blooms (mainly the genus Microcystis) frequently in summer. To improve the deteriorated water quality, a water channel was constructed in 2000 to convey a large amount of water (maximum volume: 8.6 × 10⁵ m³ day⁻¹) from the Tone River to Lake Tega. After water conveyance, the highest total nitrogen (TN) concentration in Lake Tega has decreased to 2.2 mg-N L⁻¹ in the past five years from 5.3 mg-N L⁻¹ in the 1990s. Simultaneously, the highest concentration of total phosphorus (TP) in the past five years has dropped to nearly one-half of the highest value in the 1990s (Chiba Prefectural Government 2019). The dominant algal species changed into diatoms (mainly the genus Cyclotella) in summer instead of Microcystis, which reduced the frequency of Microcystis blooms' occurrence (Sugimoto et al. 2016).

To explore the effect of water transfer on the growth and dominant shift of Microcystis and Cyclotella, Sugimoto et al. (2016) have concluded that the dilution rate of up to 36% per day or more was effective to control M. aeruginosa growth according to the result of competition experiments with Cyclotella sp. However, the inhibitory effect of water transfer on the growth of M. aeruginosa was still not considered when the dilution rate was lower than 36%. Afterward, domination characteristics for each species under various nitrate-nitrogen (NO₃-N): phosphate-phosphorus (PO₄-P) mass ratios and dilution rates were predicted by Mikawa et al. (2017) based on their algal growth simulation model. The effective dilution rate to suppress the growth of M. aeruginosa was proposed at 20% per day by a simulated result in the study. However, the predicted cell densities were always overestimated compared with experimental values due to the defect of the simulation model. In order to modify the overestimation problem in their models, Chujo et al. (2021) introduced an additional growth limitation term from Lotka-Volterra model into the previous model to develop an improved simulation model. The developed simulation model could present a more accurate prediction of the growth patterns of Microcystis sp. and C. meneghiniana, which were isolated from Lake Tega, in competition experiments under high nitrogen and phosphorus concentrations.

Although Chujo's model has been confirmed to work under eutrophic nutrient concentrations (Chujo et al. 2020), the accuracy of the model simulation has been still unclear under the condition with the combination of one nutrient being limited and another one being enough. Maberly et al. (2020) have estimated the nutrient concentrations of 17 freshwater meres in Shropshire and Cheshire, England. They found that some meres contained high nitrogen and limited phosphorus, while the nutrient concentration conditions in some other meres were opposite. Takamura et al. (1992) reported that the concentration of nitrogen exceeded that of phosphorus by 20 times after 1986 in Lake Kasumigaura, Japan. The dominant algae had shifted from Microcystis spp. to Planktothrix agardhii at that time, which was related to the phosphorus limitation and the disparity between nitrogen and phosphorus concentrations. For revealing a more accurate dilution rate that can effectively suppress Microcystis blooms in such actual lakes, it should be examined whether Chujo's model is practical under various combinations of nutrient concentration. Otherwise, the predicted values by the model would not be applicable to Microcystis bloom suppression.

In this study, we aimed to explore the mechanism of suppressive effect of dilution on the growth of two different species and examine the broad applicability of Chujo's model, then indicate an effective dilution rate to suppress Microcystis blooms in Lake Tega. For exploring the mechanism of dilution's suppressive effect, a competition experiment of two algal species, Microcystis sp. and C. meneghiniana, was conducted in phosphorus-limited and nitrogen-enough conditions under various dilution rates. Based on the model, the growth curves were simulated, and the accuracy of simulated data was tested by comparing with the experimental data. After the accuracy was verified, the growth of the two algal species under simultaneous changes of various nitrogen, phosphorus concentrations, and daily renewal rates (used instead of dilution rate in semi-continuous culture system) was simulated based on the model. The simulated values could evaluate an effective dilution rate which suppresses Microcystis blooms effectively under various nitrogen and phosphorus concentrations. The outcome was compared with the field observation of Microcystis growth in Lake Tega under different dilution rates to discuss the optimum way to suppress Microcystis blooms.

Microcystis sp. and C. meneghiniana in Lake Tega water were isolated (Chujo et al. 2019) and used as test algae in this study. The average single cell volume of the two species was measured to be 9.73 μm³ for Microcystis sp. and 243.35 μm³ for C. meneghiniana. Wright's cryptophytes (WC) medium (Guillard & Lorenzen 1972) at pH 8.0 that can cultivate both cyanobacteria and diatoms (Anderson et al. 2005) was selected as culture medium. Sodium nitrate (NaNO₃) and dipotassium hydrogen phosphate (K₂HPO₄) were used to adjust the initial nitrate-nitrogen (N) and phosphate-phosphorus (P) concentrations of the culture medium to be 14 mg-N L⁻¹ and 1.55 mg-P L⁻¹, respectively. Silicate-silicon (Si) concentration in the medium supplied by sodium metasilicate (Na₂SiO₃) was raised from 2.8 mg-Si L⁻¹ to 11 mg-Si L⁻¹, which is the same as the concentration in the Tone River water. Carbon source was provided by sodium hydrogen carbonate (NaHCO₃). The medium was autoclaved at 121 °C for 20 minutes in all experiments.

For subculture, Microcystis sp. and C. meneghiniana were separately cultivated in WC medium in a 300 mL Erlenmeyer flask. Each flask was incubated in an incubator (MTI-202, EYELA, Japan) at a temperature of 25 °C, and the light intensity was adjusted to 10,000 lux using a light meter (LX-1102, Lutron, America). The light-dark cycle was set as 14 hours-light and 10 hours-dark. The two subcultured species were inoculated to fresh medium every two or three weeks. All the operations of inoculation and sampling were conducted in a clean bench to minimize bacterial contamination.

Semi-continuous competition experiments were conducted with various dilution rates under the limited phosphorus and enough nitrate concentrations, to investigate the combined effect of these conditions on Microcystis sp. and C. meneghiniana.

Prior to the competition experiment, both species were precultured in nitrogen- and phosphorus-free medium for 7 days to deplete intracellular N and P, thereby eliminating the effect of intracellular nutrients. After the preculture, both species were inoculated together in 200 mL sterilized medium in a 300 mL Erlenmeyer flask for semi-continuous competition experiments. The initial cell density of Microcystis sp. was adjusted to 1.0 × 10⁴ cells mL⁻¹ and that of C. meneghiniana was 3.98 × 10² cells mL⁻¹ to make the two species' cell volume equivalent to the same volume of 10⁵ μm³.

The daily renewal rates of the competition experiment were set as 0, 5, and 15%. An appropriate volume of culture medium was removed and as soon an equal volume of fresh medium was added to each flask once a day. In addition to the different daily renewal rates, the P concentration of each group was limited to 0.1 mg-P L⁻¹, which was the same as the minimum P concentration in Lake Tega in the last five years (Chiba Prefectural Government 2019). The initial N concentration of 14 mg-N L⁻¹ was adequate for the growth of both species (Anderson et al. 2005). The cell density was measured every 2–5 days and continued until it tended to be constant. The nutrient concentrations were measured once a week. The experiments with different dilution rates were conducted in triplicate. All the experimental results were presented as [the mean value]±[standard deviation].

Concentrations of N were measured by ion chromatography (ICS-1100, Nippon Dionex, Japan), and molybdenum blue method was used to measure P concentration. The solution pH was monitored by a pH meter (D-51, Horiba, Japan).

Differences in experimental parameters of Microcystis sp. and C. meneghiniana in each condition were analyzed by a one-way analysis of variance (ANOVA) with a post hoc comparison being performed with Turkey's test, via SPSS Statistics (Ver. 23, IBM Corporation, USA). The results were judged to be a significant difference at p < 0.05.

To predict the cell densities of Microcystis sp. and C. meneghiniana when they grow together and the trends of the appearance of Microcystis blooms under various nutrient concentrations and daily renewal rates, the model constructed by Chujo et al. (2020) was used. The equations of the model are tabulated in Table 1.

The accuracy of a previous model (Mikawa et al. 2016) tended to decrease under high nutrient concentrations. In the previous model, the limiting nutrient was determined as the relationship between the mass ratio of minimum cell quota of assimilated N:P (the optimum N:P ratio) and the external dissolved N:P mass ratio. While in the Chujo's model, the relationship between the mass ratio of cell quota of assimilated N:P (Q_n:Q_p) and the optimum N:P ratio was taken to determine the limiting nutrient (as shown in the tag of Table 1).

The model equations were calculated via a fourth-order form of the Runge-Kutta method with the time step of Δt = 0.01 day, using Microsoft Excel. The values of each constant parameter in the equations taken to calculate were shown in Table 2, which were obtained from monoculture experiments in the study of Chujo et al. (2021). Furthermore, to investigate the accuracy of the model simulation under combined nutrient concentrations, the growth curves of both species were simulated under the same experimental condition as the semi-continuous competition experiments mentioned above.

In the investigation of effective dilution rate, the growth of both species was simulated under various nutrient concentrations and dilution rates based on the Chujo's model. The initial N and P concentrations were ranged from 0 to 5.0 mg-N L⁻¹ and 0 to 0.5 mg-P L⁻¹, respectively, reflecting the nutrient concentration in Lake Tega. The daily renewal rate (d) was increased from 0 to 20% with a 2.5% interval for each step. Since the two species always reached saturation within around 20 days (Chujo et al. 2021), the cell density in day 30 was used as the final result for prediction.

Growth characteristics for Microcystis sp. and C. meneghiniana in P limited culture experiment are shown in Figure 1. The simulated growth curves of both species are also depicted in the same figure.

In all daily renewal rates (d) experimental groups, the biovolume of both species has reached saturation within 25 days. At d = 0%, the biovolume of Microcystis sp. was 8.13 × 10⁶ ± 1.76 × 10⁶ μm³ L⁻¹ at day 24, and after that, the biovolume remained constant within 7 days. For the other two experimental groups (d =5 and 15%), there was no significant discrepancy in the biovolume (p >0.05). In the d = 5% group, the biovolume in saturation was obtained at day 17 as 9.31 × 10⁶ ± 1.29 × 10⁶ μm³ L⁻¹ and that for the 15% group was 1.31 × 10⁷ ± 1.17 × 10⁶ μm³ L⁻¹ at day 21. The average growth rate of Microcystis sp. in the logarithmic phase of the three groups was also similar (p >0.05). The value was 0.313 ± 0.007 day⁻¹ in the d = 0% group, and 0.318 ± 0.005 day⁻¹ in the d = 5% group. When the daily renewal rate increased to 15%, the average growth rate in the logarithmic phase was obtained to be 0.291 ± 0.002 day⁻¹.

On the other hand, the growth of C. meneghiniana was affected obviously by the daily renewal rate under phosphorus limitation. C. meneghiniana reached saturation at day 7 with the biovolume of 3.76 × 10⁷ ± 1.01 × 10⁶ μm³ L⁻¹ in the d = 10% group and remained stationary until the end of the experiment. In comparison, the growth of C. meneghiniana at d = 15% has entered the death phase directly after it reached saturation at day 5. Moreover, the biovolume in this group at the end of the experiment was 2.18 × 10⁶ ± 1.01 × 10⁶ μm³ L⁻¹, which decreased by nearly one order of magnitude compared with that in the d = 5% group. As a result of the competition experiments, the dominant species, defined as the species which take the most biovolume in the flask, was always Microcystis sp. in each group. The obvious effect of the daily renewal rate was also reflected in the growth rate of C. meneghiniana. At the beginning of the experiment (from day 0 to day 5), although the differences in growth rate among each group were not obvious (p >0.05), the average growth rates of the three groups displayed different characteristics after the logarithmic phase. From day 12 to day 18, the average growth rate in the d = 0% group was 0.005 ± 0.001 day⁻¹, and then the growth rate of C. meneghiniana gradually changed to 0 until the end of the experiment. On the contrary, in the groups of d = 5% and d = 15%, the average growth rate has become negative after the logarithmic phase.

From the result of nutrient analysis as shown in Figure 2, it was observed that the dilution had scant effect on the change in nutrient concentrations in the competition experiments. The differences in the N and P concentrations among each group were not significant (p >0.05), and the decline of both nutrient concentrations was mainly attributed to the growth of two species as the competition experiment progressed. The concentration of P dropped rapidly during the logarithmic phase of the two species and remained at 0.012 ± 0.002 mg-P L⁻¹ and 0.016 ± 0.004 mg-P L⁻¹ at d = 0% and d = 15%, while the N concentration remained at 9.25 ± 0.39 mg-N L⁻¹ and 8.62 ± 0.22 mg-N L⁻¹ at d = 0% and d = 15%, respectively. These values portrayed that the nutrient concentrations were similar at the end of the competition experiments, despite the differences in daily renewal rates.

The simulated growth curves in Figure 1 displayed that the simulated data could well correspond with the experimental growth patterns for Microcystis sp. In the d = 0% group, the trend of experimental cell growth data and simulated curve changing trend have coherence. The two simulated growth curves in the d = 5% and d = 15% group almost coincided with each other, which is consistent with the fact that there was not a significant difference in experimental data between the two groups (p >0.05). The coefficient of determination (R²) for the simulated curve at d = 0% was 0.827 and the values at d = 5% and d = 15% were 0.894 and 0.915, respectively. Thus, the feasibility of Chujo's model under limited phosphorus and sufficient nitrogen conditions could be verified. It would be used to predict the growth of Microcystis sp. under various nutrient conditions including phosphorus deficiency. Similarly, the simulated values also matched with the experimental patterns in the growth of C. meneghiniana at d = 0% under low P concentration (R² = 0.973). Although a little discrepancy between simulated and experimental biovolume was observed at d = 5 and 15%, the model can still be valid to simulate the growth trend of C. meneghiniana.

Since the accuracy of the model was verified, the cell density of Microcystis sp. was predicted under the same N and P concentrations and daily renewal rate as those in Lake Tega before (d = 5%) and after (d = 15%) water transfer. The estimated Microcystis sp. cell densities in saturation were 2.72 × 10⁶ cells mL⁻¹ (before water transfer) and 1.36 × 10⁶ cells mL⁻¹ (after water transfer), respectively. Based on these values, it could be assumed that, when the estimated cell density was more than 2.72 × 10⁶ cells mL⁻¹, there is a high possibility of Microcystis blooms occurrence, while the cell density of less than 1.36 × 10⁶ cells mL⁻¹ implies that blooms would occur hardly. The plane view of contour figures in Figure 3 shows the predicted values at various daily renewal rates and different nutrient concentrations. The black area was taken as the bloom occurrence area where the predicted cell densities were equal to or more than 2.72 × 10⁶ cells mL⁻¹. The dark-grey area was taken as the non-bloom occurrence area where the predicted cell densities were equal to or less than 1.36 × 10⁶ cells mL⁻¹. The light-grey area between the black and dark-grey area implies the state that suffers the risk of bloom appearance although the blooms have not still occurred. Solid lines and dashed lines were used to indicate the boundary of the bloom occurrence area and non-bloom occurrence area, respectively.

When the daily renewal rate changed from 0 to 5%, the bloom area was obviously enlarged. However, as the daily renewal rate gradually increased from 5%, the bloom area gradually decreased. At the same time, the non-bloom occurrence area increased. It could be anticipated that the Microcystis bloom area would disappear at a certain dilution rate. In particular, it can be found in the figures that when the renewal rate was equal to or higher than 12.5%, the bloom area no longer existed. Based on Equation (1), the dilution rate D was 13.3% when the renewal rate was 12.5%.

The data from Chiba Prefectural Government (Chiba Prefectural Government 2019) was used to calculate the annual average of dilution rate (D_avg) and the cell densities of Microcystis sp. in every month of each year in Lake Tega. The results are displayed in Figure 4. The field observation data indicated that when the dilution rate was above 13.3%, Microcystis sp. was hard to proliferate, and the cell density was less than 3 × 10³ cells mL⁻¹. This trend apparently corresponds to the predicted data as shown in Figure 3. Thus, the dilution rate D = 13.3% could be considered to suppress Microcystis blooms effectively.

Semi-continuous competition experiments showed that the growth of C. meneghiniana isolated from Lake Tega was more affected by the replacement of medium (dilution) than that of Microcystis sp. under phosphorus limitation. Meanwhile, the dilution still had a suppressive effect on the growth of Microcystis sp. The simulation results for the competitive growth elucidated that the lowest dilution rate to suppress Microcystis blooms was D = 13.3%, which is consistent with the field observation results in Lake Tega.

In semi-continuous competition experiments, Microcystis sp. showed better adaptability than C. meneghiniana under the phosphorus limited condition. The growth parameters obtained from monoculture experiments of the two species in the previous study (Chujo et al. 2021) as shown in Table 2 indicate that the maximum cell quota (Q_max) values of C. meneghiniana for N and P (Q_max,n = 104 pg cell⁻¹, Q_max,p = 5.08 pg cell⁻¹) were nearly 20 times higher than those of Microcystis sp. (Q_max,n = 5 pg cell⁻¹, Q_max,p = 0.28 pg cell⁻¹), whereas the uptake rate ρ_max^hi for N was only twice that of Microcystis sp. This indicates that the deficiency of phosphorus would influence the growth of C. meneghiniana more than Microcystis sp., and that Microcystis sp. would grow more advantageously with adequate nutrients, excluding the effect of other factors such as temperature and dilution. Based on the changes of nutrient concentration as shown in Figure 2, there should be no remarkable differences in cell densities of both species in each group. However, C. meneghiniana cell densities decreased sharply with the increase in daily renewal rate. Thus, it could be considered that the number of C. meneghiniana cells washed away everyday was greater than that grown under the phosphorus limitation. Furthermore, the fluctuation of phosphorus concentration caused by water transfer has been proven that it can not lead to the shift of dominant species and suppress the formation of Microcystis blooms (Amano et al. 2010). This was consistent with our results in the semi-continuous competition experiment.

Tatsumoto et al. (2008) indicated that the ignition loss of the Lake Tega sediment was measured to be ca. 16% and the upper 20 cm of the sediment was supposed to contribute to nutrient elution, which may provide nutrient constantly after water transfer. Water pH is also an important influential factor of the nutrient release from sediment, although the release of ammonium-nitrogen increases with pH and the release of nitrate-nitrogen has no obvious relation with pH (Hu et al. 2003). The release of phosphorus will decrease with increasing pH until 7.0, and increase with rising pH after higher than 8.0 (Wu et al. 2014). Furthermore, the contribution of different ratios of various nitrogen sources to the growth of Microcystis sp. has also been observed by several studies (Junfeng et al. 2017; Krausfeldt et al. 2020). In the north part of Lake Taihu, China, when the molar ratio of ammonium to nitrate was below 1, Microcystis blooms tended to be dominant during summer (Liu et al. 2011). Therefore, methods of controlling the internal pollution, adjusting the nutrient concentrations, and controlling the pH of the water body were essential for the prevention and management of cyanobacterial (Microcystis) blooms and should be developed in the future.

As for the simulated results shown in Figure 3, the prediction of Microcystis sp. growth displayed a decrease at daily renewal rates of lower than 5%, and it was very nearly impossible for the cyanobacteria blooms to occur when the daily renewal rate was 0%. However, in reality, the smaller the daily renewal rate is, the more possibility the cyanobacterial blooms occurrence should be (Romo et al. 2013). The discrepancy would be due to a shortcoming with the nutrient uptake term of the Chujo's model. According to the equations of the model (Table 1), the value of the uptake rate term ρ_max was always obtained when the nutrient concentration was adequate. However, it could not always reach the maximum value and would decrease with algal growth as well as nutrient concentration decline in the competition process. As a result of the shortcoming, the simulated nutrient concentrations always decreased exponentially when the cell densities of Microcystis sp. increase exponentially. In comparison, in reality, when the cell density increased exponentially with initial sufficient nutrients, the decrease in nutrient concentration was only proportionally reduced (Wang et al. 2010; Chujo et al. 2021). For this reason, the predicted nutrients uptake of Microcystis sp. after the logarithmic phase would be overestimated.

However, when the two species were grown together under eutrophic conditions, C. meneghiniana showed a very advantage in nutrient uptake over Microcystis sp. as shown in our previous study (Chujo et al. 2021). At the daily renewal rate of 0%, both the experimental and the simulation results indicated that the nutrient was almost exclusively occupied by C. meneghiniana, which led to the suppression of the proliferation of Microcystis sp. This mechanism provided a reasonable explanation and support for the predicted cell density of Microcystis sp. at d = 0%. Nevertheless, there was still a gap between the experimental and predicted values of the C. meneghiniana cell densities when it was under relatively scarce nutrient concentration. With the objective of improving the accuracy of the simulation model, the amelioration of the uptake rate term and other environmental factors such as temperature and light intensity should be focused on in the future study.

The replacement water (medium) restricted both the growth of C. meneghiniana and Microcystis sp. in competition experiments. Especially, the restricted effect was more significant on the growth of C. meneghiniana compared with that of Microcystis sp. under a nutrient condition with the combination of limited P and sufficient N. The difference was caused by the more excellent ability of nutrient uptake of Microcystis sp., and C. meneghiniana was easier to flow away. The broad accuracy of Chujo's competition growth model in various nutrient concentrations was also verified based on the prediction curves and competition experiments results. The minimum effective dilution rate (D) to suppress Microcystis sp. bloom in Lake Tega was estimated to be 13.3% by this model, which was consistent with the field observation in Lake Tega. Since the accuracy of the Chujo's model in broad nutrient concentration range has been verified, it can be a powerful tool to apply for the management of Microcystis bloom in many actual lakes with complex nutrient conditions.

This work was supported in part by the Japan Society for the Promotion of Science under Grants-in-aid for Scientific Research (No. JP21K04311) and by the JST SPRING (No. JPMJSP2109). Authors are thankful to Prof. Dr. Fumio Imazeki, the head of Safety and Health Organization at Chiba University, for his encouragement and financial support.

All relevant data are included in the paper or its Supplementary Information.