Simulation study on algal dynamics based on ecological flume experiment in Taihu Lake, China

Simulation study on algal dynamics based on ecological flume experiment in Taihu Lake, China

e c o l o g i c a l e n g i n e e r i n g 3 1 ( 2 0 0 7 ) 200–206 available at www.sciencedirect.com journal homepage: www.elsevier.com/locate/ecole...

690KB Sizes 0 Downloads 16 Views

e c o l o g i c a l e n g i n e e r i n g 3 1 ( 2 0 0 7 ) 200–206

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/ecoleng

Simulation study on algal dynamics based on ecological flume experiment in Taihu Lake, China Ling Ding a,∗ , Jian Q. Wu b , Yong Pang c , Ling Li d , Guang Gao e , Dong W. Hu b a

Shanghai Investigation, Design & Research Institute, Shanghai 200434, China Shanghai Academy of Environmental Sciences, Shanghai 200233, China c Institute of Environmental Science and Engineering, Hohai University, Nanjing 210098, China d Environmental Modelling School of Engineering, The University of Queensland, Brisbane, Qld 4072, Australia e Nanjing Institute of Geography and Limnology, Chinese Academy of Sciences, Nanjing 210008, China b

a r t i c l e

i n f o

a b s t r a c t

Article history:

The dynamic growth of algae is influenced by the inner physiological characteristics as well

Received 12 July 2005

as the exterior stimulating factors, including light, temperature and nutrients. The sus-

Received in revised form 4 May 2007

pended sediments release nutrients driven by hydrodynamic forces, imposing significant

Accepted 25 June 2007

effects on algae growth. In this research, the mechanism of algae growth was studied, and an experiment on algae growth was conducted in an ecological flume using water and sediments collected from Taihu lake, an eutrophic lake in China, with three water current stages:

Keywords:

still water stage, slow water current stage, and fast water current stage. Based on the exper-

Hydrodynamic forces

iment, a model was developed to quantitatively describe the relations between the release

Ecologic flume experiment

of nitrogen/phosphorous nutrients and the water flow velocity in nitrogen and phospho-

Algae growth model

rus cycles. Detailed calibration and verification of the model parameters were carried out

Taihu Lake

based on the results from experiments. The predicted values fit presented the results fitting well with the measured data, which indicates that the developed model can represent the dynamics of algae growth to a certain extent. This model has the potential for forecasting algal bloom in shallow, temperate lake systems. © 2007 Elsevier B.V. All rights reserved.

1.

Introduction

Taihu Lake, located near Wuxi, China (shown in Fig. 1), is a highly eutrophic lake. Since the end of the 1980s, the blue algae bloom has frequently occurred in Taihu Lake. Algal bloom outbreak makes the water body smell fishy, decreases water transparency, exhausts dissolved oxygen, and releases poisonous substances, which greatly imperils industrial and drinking water, aquatic ecology, and fishery. Especially in 1990, the algae density of Meiling Bay in Taihu Lake maintained a high magnitude for about 25 days. The waterworks were forced into reduction in yield and nearly 100



Corresponding author. Tel.: +86 21 65427100. E-mail address: dingling [email protected] (L. Ding). 0925-8574/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.ecoleng.2007.06.013

industries had to reduce or stop production; the dead fishes in Taihu Lake weighed 45,000 kg. The economic losses caused by this algae bloom were higher than one hundred million yuan. In short, exceptionally rapid growth (blooms) of algae has been a great environmental problem in Taihu Lake. The study of lake eutrophication has been a hot topic in the world since the 1980s. Many scholars are focusing on algal kinetics research, such as experimental study and simulation study (Colleen et al., 2005; Eran et al., 2003; Leah et al., 2005; Richard et al., 2005; Sandra et al., 2002; Sebnem and Ilgi, 2006). In China, as for the mechanism study of algae growth in Taihu Lake, Yuanbo (1998) and Qiujin (2001) had developed a model

e c o l o g i c a l e n g i n e e r i n g 3 1 ( 2 0 0 7 ) 200–206

201

Fig. 1 – The location of Taihu Lake in China.

on algae growth in Taihu Lake. In their model, they only considered the phosphorus factor on algae growth without the nitrogen factor. In addition, the hydrodynamic factor was not taken into account. Taihu Lake is a large shallow lake with an average depth of less than 2 m. Studies have indicated that hydrodynamic forces impose significant effects both on the suspension of sediments and nutrients released, contributing to algae growth in Taihu Lake (Boqiang and Chengxin, 2002). Accordingly, hydrodynamic force is an important factor for algae growth. Additionally, in their model, the on-site monitoring data were used to simulate the algae growth. As the monitoring proceeded at each site unsynchronous, the precision of model was affected. In this study, a dynamic simulation experiment is conducted in an ecological flume aiming to discover the mechanism of algae growth. Based on the experiment, an algae growth model was developed. In this model, nutrients release due to hydrodynamics forces, contributing to algae growth, are considered in nitrogen and phosphorus cycle. The quantitative relations between the release rate of nitrogen/phosphorus nutrients derived from sediment and the water flow velocity were studied through the experiment conducted in an annular flume, and incorporated in the model. The model is used to simulate algal growth in an ecological flume experiment and to describe the mechanism of algae growth quantitatively. The model has the potential to play a role in forewarning of the outbreak of algal blooms.

2.

Ecological flume experiment

The ecological flume experiment was conducted from 8th May to 24th June in 1999. The ecological flume was 6.5 m long, 1.3 m wide and 1.6 m high, as shown in Fig. 2. Sediments collected from Taihu Lake were laid uniformly on the bottom of the flume. The sediment thickness was set to 10 cm. Then water from the lake was infused into the flume and the water depth was set to 1.15 m. The experiment was conducted in an airconditioned laboratory where the air temperature was kept at 25 ◦ C. Light source was the sunlight. Water temperature and light intensity was measured once a day. The experi-

ment was divided into three phases: still water stage, slow water current stage, and fast water current stage. The slow water current stage was started on 23rd May and was run continuously for 10 days with a constant average flow velocity of 0.124 m/s. Then water samples were collected regularly for 5 days, during which period the slow water current state remained unchanged. The fast water current stage was started on 8th June and was similarly run for 10 days with a constant average flow velocity of 0.319 m/s. Water samples were also collected regularly for 5 days under the fast water current condition. Light intensity was measured by an underwater optical photon instrument produced by LI-COR Corporation in USA. The instrument can accurately measure the intensity of available, photosynthetic radiance (PAR, 400–700) even in turbid water where the light intensity is less than 0.01 ␮mol/m2 s−1 . Chlorophyll-a concentration was measured by a spectrophotometer. The chlorophyll-a concentration can be derived according to [Chl-a] = (27.9(Eb − Ea )Ve )/V where [Chl-a] is the chlorophyll-a concentration; Ea the spectral density difference between 665 and 750 nm wavelength before acidification; Eb the spectral density difference between 665 and 750 nm wavelength after acidification; Ve the volume of extracting solution; and V is the volume of water sample delivered by a suction strainer. Flow velocity was measured by a LS78 type lowspeed tachometer. The slow water current was controlled by a QDX10-10-0.55 sinking pump and the fast water current was controlled by a QY oil immersion type sinking pump. Nutrient concentrations were measured using the method of Xiangfei (1999).

Fig. 2 – The schematic diagram of the ecological flume.

202

e c o l o g i c a l e n g i n e e r i n g 3 1 ( 2 0 0 7 ) 200–206

Fig. 3 – The relations between algae and nutrients described in model.

3.

Taihu algal growth model

It is generally considered that the ecological processes of algae are mainly related to light, temperature, nutrients as well as the algae’s physiologic factors. In this study, the algal growth model is shown in Fig. 3. Six variables are considered in the model, which are algal biomass, ammonium, nitrate, organic nitrogen, inorganic phosphorus and organic phosphorus, respectively. The relations between these variables are as follows: ammonium, nitrate and inorganic phosphorus are taken up by algae, and meanwhile, parts of algal nitrogen/phosphorus are converted into organic nitrogen/phosphorus. By mineralization, organic nitrogen/phosphorus is converted into ammonium/inorganic phosphorus, respectively. Ammonium is ultimately converted into nitrate by nitrification. In addition, release of nutrients from the sediment driven by the water current is considered in the nitrogen/phosphorus cycle. Using the algae growth model, the dynamic processes of algae during two different water current regimes are simulated. By calibration and verification with the experimental measured data, we aim to develop the quantitative relations between physico-chemical factors and algae growth.

3.1.

Model description

The governing equations of the model are developed as follows: ∂C1 = ∂t



Gp1 − Dp1 −

Vs1 D



C1

(1)

Vs3 ∂C2 T−20 = Dp1 C1 apc − K23 23 (1 − fD2 )C2 + S2 C2 − ∂t D

(2)

∂C3 Vs5 T−20 C2 − Gp1 C1 apc − = K23 23 (1 − fD3 )C3 + S3 ∂t D

(3)

Vs3 ∂C4 T−20 = Dp1 C1 aNC − K45 45 C4 − (1 − fD4 )C4 + S4 ∂t D

(4)

∂C5 T−20 T−20 C4 − Gp1 aNC C1 PNH3 − K56 56 C5 + S5 = K45 45 ∂t

(5)

∂C6 T−20 = K56 56 C5 − Gp1 aNC C1 (1 − PNH3 ) + S6 ∂t

(6)

Gp1 = max f (T)f (I)f (N)f (P)

(7)

T−20 f (T) = K1c 1c

(8)

f (I) =

I KI + I

(9)

f (N) =

N KN + N

(10)

f (P) =

P KP + P

(11)

T−20 + K1D Dp1 = K1R 1R

(12)

where C1 is the algal biomass, Cp1 the growth rate of algae, which is limited by the factors of temperature, light and nutrients; T the temperature, I the radiant intensity, N and P the concentration of nitrogen and phosphorus, respectively; Dp1 the loss rate of algae, including respiration rate and death rate; Vs1 /D the sedimentation of algae, D the water depth; C2 the concentration of organic phosphorus, Dp1 C1 apc the term of conversion of algal phosphorus to organic phosphorus, T−20 K23 23 C2 the mineralization term of conversion of organic phosphorus to inorganic phosphorus, (Vs3 /D)(1 − fD2 )C2 the sedimentation of organic phosphorus, S2 the release rate of organic phosphorus from the sediment; C3 the concentration of inorganic phosphorus, Gp1 C1 apc the uptake of phosphorus by algae, (Vs5 /D)(1 − fD3 )C3 the sedimentation of inorganic phosphorus, S3 the release rate of inorganic phosphorus from the sediment; C4 the concentration of organic nitrogen, Dp1 C1 aNC the term of conversion of algal nitrogen to T−20 organic nitrogen, K45 45 C4 the mineralization term of conversion of organic nitrogen to ammonium, (Vs3 /D)(1 − fD4 )C4 the sedimentation of organic nitrogen, S4 the release rate of organic nitrogen from the sediment; C5 the concentration of

203

e c o l o g i c a l e n g i n e e r i n g 3 1 ( 2 0 0 7 ) 200–206

ammonium, Gp1 aNC C1 PNH3 the uptake of ammonium by algae, T−20 C5 the nitrification term of conversion of ammonium K56 56 to nitrate, S5 the release rate of ammonium from the sediment; C6 the concentration of nitrate, Gp1 aNC C1 (1 − PNH3 ) uptake of nitrate by algae, and S6 the release rate of nitrate from the sediment. As hydrodynamic force drives the sediment suspension resulting in the nutrients release, the release rate of nutrients is considered presenting the relationship with water flow velocity in this model. To determine such relations (between release rate and flow velocity), the experiment on nutrient release was conducted in a circulating flume at the College of Coastal and Ocean Engineering, Hohai University from 26th to 30th April in 2003. First, sediment of Taihu Lake was paved evenly on the bottom of the flume. The sediment thickness was set to be 5 cm. Fresh water of Taihu Lake was infused into the circulating flume lightly by siphon. The water depth was 15 cm. Then after the water and sediment were kept static for 24 h, the flow velocity remained at 0, 5, 8.5, 17.5, 25, 30, 40, 50 and 60 cm/s, respectively, by microcomputer controller. The flow velocity was increased gradually from 0 to 60 cm/s. At each velocity, samples were taken from upper layer, mid layer and bottom layer of water (detailed in Yiping et al., 2004). During the experiment, the sediment suspension experienced three stages: individual, ounce and universal movement. Accordingly, nutrients in the sediment were released to the upper water. The formula for release rate of nutrients is as follows (Shuncai and Yiping, 1993): r=

¯ n − c0 ) + V(c

n

j=1

Numerical methods

A fifth-order Runge–Kutta algorithm is used to integrate the above six partial differential equations. For example, Eq. (1) can be rewritten as follows: dC1 = F(C1 , C2 , C3 , C4 , C5 , C6 , t) dt This equation can be integrated with the following initial conditions using the Runge–Kutta algorithm: C1 = C10 , C2 = C20 , C3 = C30 , C4 = C40 , C5 = C50 , C6 = C60 , at t = 0

The integrated form of Eq. (1) is then C1i+1 = C1i + tϕ(ti , C1i , C2i , C3i , C4i , C5i , C6i , t) Where ϕ(ti , C1i , C2i , C3i , C4i , C5i , C6i , t) =

K1 = F(ti , C1i , C2i , C3i , C4i , C5i , C6i )

⎛ Ki = F ⎝ti + i t, C1i + t

At

yTP = 36.78 exp(0.05v),

R2 = 0.94; R2 = 0.97

where y is the release rate of TN or TP from the sediments [mg/m2 day−1 ],  the water flow velocity (cm/s), and R is the regression coefficient. The study indicated that the release rates for different forms of nitrogen nutrients differ slightly in the relation with the flow velocity: the rate constants present different values. The same variations exist for TP and different forms of phosphorus nutrients (Chengxin et al., 1998). Such variations are incorporated in the model, i.e., (i = 2, 3);

Si = ai exp(0.06v),

(i = 4, 5, 6)

The values of coefficient ai are obtained by calibration on the basis of the results from the ecological flume experiment.

ci Ki

i=1

Vi (cj−1 − ca )

yTN = 137.88 exp(0.06v),

5 

i−1 

ij Kj , C2i + t

j=1

where r is the release rate of nutrients (mg/m2 day−1 ); V the volume of water in circulating flume (L); cn the nutrients concentration of the sample n taken at nth time (mg L−1 ); c0 the initial nutrients concentration (mg L−1 ); Vi the volume of sample (L); cj − 1 the nutrients concentration of sample (j − 1) taken at the (j − 1)th time (mg L−1 ); Ca the nutrients concentration when raw water is supplied (mg L−1 ); t the release duration (day); and A the surface of sediment contacted with water (m2 ). According to the experimental results and the formula for release rate, the quantitative relations between the release rate of nutrients from the sediment and the flow velocity have been developed as

Si =ai exp(0.05v),

3.2.

+t

i−1  j=1

+t

i−1 

ij Kj , C4i + t

⎞ ij Kj ⎠

i−1 

i−1 

ij Kj , C3i

j=1

ij Kj , C5i + t

j=1

i−1 

ij Kj , C6i

j=1

i = 2, . . . , 5

j=1

Ci , I and ij are all constants.

4.

Results and discussion

4.1.

Experiment results

Table 1 shows the range and average values of the physicochemical factors measured during the two water current stages. From Table 1, we can find that the concentrations of ammonium, nitrate, organic nitrogen, inorganic phosphorus, and organic phosphorus under two different water current regimes showed a similar trend, with higher values in fast water current regime. This indicated that with increasing flow velocity, more sediment suspended in the water with nutrients in it was released accordingly. When nutrient contents in water were increasing, the algae had more chances to obtain nutrients for growth. So the algal biomass was larger in fast water current stage than in slow water current stage. In addition, Table 1 shows that the increase in water current is followed by sediment suspension with the decreasing water transparency from 0.65 to 0.25 m, which caused the attenuation of underwater light intensity. In general, the hydrody-

204

e c o l o g i c a l e n g i n e e r i n g 3 1 ( 2 0 0 7 ) 200–206

Table 1 – The range and average values of physico-chemical factors during two different water current stages Algal biomass (cell/L) Range Slow water current Fast water current

Average

1.6 × 10 to 4.8 × 10 5.1 × 105 to 6.2 × 107 6

6

Slow water current Fast water current

0-5 3-20

Range

2.9 × 10 3.4 × 107

Inorganic phosphorus (␮g/L) Range

Ammonium (␮g/L)

Average 2.7 10.4

6

Average

0-11 0-49

Range

6 11

Organic nitrogen (mg/L)

Average

0.516-0.669 0.841-1.041

Organic phosphorus (␮g/L)

0.577 0.963

Range 0.645-0.723 0.945-1.284

Underwater light intensity (␮mol/m2 s−1 )

Average 0.692 1.132

Water transparency (m)

Velocity (m/s)

Range

Average

Range

Average

Average

Average

36-54 86-104

40 97

178-395 92-281

316 207

0.65 0.25

0.124 0.319

namic forces changing the nutrient concentrations and underwater light intensity result in the variation of algal biomass.

4.2.

Nitrate (mg/L)

Simulation of algal dynamics

Based on experimental results, the dynamic process of algae growth under slow flow current stage is simulated by algae growth model. Model parameters were adjusted with the difference between calculated and measured values in allowable error range and therefore some groups of parameters were obtained. The least square method was used to determine the optimal parameter group. Then the experimental results of the fast water current stage are used for verification to ensure the reliability of determined parameter values. Table 2 shows the ultimately determined parameter values of the algae growth model. Fig. 4 shows the comparison

between the calculated values by modelling and the measured values in slow water current stage. Fig. 5 shows the comparison between the calculated values by modelling and the measured value in fast water current stage. As shown in Table 2, the most important parameters of algae growth model such as max , KN , KP , K1R determined in this study are all in the range of values found in the literature. The values of K1C , KI , K1D , apc , aNC , K45 are all close to the studied results of Yong (1998) and Qiujin (2001) in Taihu Lake. The slight differences between some parameter values in this study and from the literature are probably due to the different ecosystem characteristics, different model structure and different data. In addition, it was noticed that the rate constant for nitrate release a6 is negative. This “negative release” phenomenon is consistent with the experimental results obtained by Chengxin et al. (1998). An explanation for

Table 2 – Parameter values of algae growth model Parameter max KN KP K1C  1C KI K1R  1R K1D apc K23  23 aNC K45  45 PNH3 a2 a3 a4 a5 a6

Meaning Maximum growth rate of algae Half saturation constant of N-uptake Half saturation constant of P-uptake Constant or coefficient at 20 centigrade degree Temperature dependence coefficient Light saturation parameter Respiration rate of algae Temperature coefficient Death rate of algae content of phosphorus in algae Rate coefficient for mineralization Temperature coefficient Content of nitrogen in algae Rate coefficient for mineralization Temperature coefficient Ratio of ammonium uptake in nitrogen uptake by algae Coefficient for release of organic phosphorus from the sediment Coefficient for release of inorganic phosphorus from the sediment Coefficient for release of organic nitrogen from the sediment Coefficient for release of ammonium from the sediment Coefficient for release of nitrate from the sediment

Unit (24 h)−1 mgL−1 mgL−1 (24 h)−1 – ␮mol/m2 s−1 (24 h)−1 – (24 h)−1 – (24 h)−1 – – (24 h)−1 – –

Value 1.0 0.14 0.015 2.0 1.12 300 0.45 1.11 0.2 0.025 0.08 1.05 0.25 0.05 1.02 0.68

Literature value 1–5 (Jørgensen, 1994) 0.05–0.5 (Jørgensen, 1994) 0.005-0.03 (Jørgensen, 1994) 2.0 (Yong, 1998) 1.068 (Yong, 1998) 300 (Qiujin, 2001) 0.1-0.5 (Jørgensen, 1994) 0.17 (Qiujin, 2001) 0.025 (Yong, 1998) 0.01-0.178 (Hongping et al., 2000) 0.8-1.08 (Hongping et al., 2000) 0.25 (Yong, 1998) 0.075 (Yong, 1998) 1.08 (Yong, 1998) -

mg (m2 day)−1

24.52

-

mg (m2 day−1 )

15.99

-

mg (m2 day)−1

7.03

-

mg (m2 day)−1

168.82

-

mg (m2 day)−1

−21.87

-

e c o l o g i c a l e n g i n e e r i n g 3 1 ( 2 0 0 7 ) 200–206

Fig. 4 – Simulated results of algae and nutrients kinetic processes in slow water current stage.

Fig. 5 – Simulated results of algae and nutrients kinetic processes in fast water current stage.

205

206

e c o l o g i c a l e n g i n e e r i n g 3 1 ( 2 0 0 7 ) 200–206

the phenomenon is that when the sediment surface lacks free oxygen, the nitrogen denitrification converts NO3 − –N into NO2 − –N, which results in the decrease of the NO3 − –N content. As time elapses, NO2 − –N continues to transfer into low valence substances such as N2 O and N2 which can escape into the atmosphere through the water-atmosphere interface. Consequently, the NO3 − –N content decreases, resulting in the “negative release” phenomenon. Figs. 4 and 5 show the simulated results of algae and nutrients kinetic processes during slow and fast water current stages, respectively. It shows that the calculated values of algae and nutrient concentrations reasonably agree well with measured values. In Fig. 4, all average relative errors between the calculated value and measured value are below 10% except for the ammonium with a relative error of 14.8%. In Fig. 5, the average relative errors between the calculated value and measured value of ammonium and inorganic phosphorus concentrations are 16.3% and 11.3%; others are less than 10%. In slow water current stage, because of the relatively lower flow velocity, and because the sediment was only disturbed slightly, no significant nutrient release occurred. The concentrations of organic nitrogen and phosphorus had no obvious changes during this stage. The concentrations of ammonium and inorganic phosphorus were related to algal biomass. On 3rd June, the algal biomass was the highest, and nutrients were taken up fast because of high algal number. Accordingly, the complete uptake of ammonium and inorganic phosphorus occurred with concentration down to 0 mg/L. Then with the decrease of algal number, ammonium and inorganic phosphorus concentrations rose up. However, the variation of nitrate concentration had no similar trend. So it is probably indicated that algae selects ammonium for uptake and has no obvious uptake demand for nitrate. In fast water current stage, with flow velocity increased, a lot of sediments suspended to the water, accompanied by fast nutrient release. The increase of nutrient contents further promoted algae growth. So the algal biomass rapidly went up at the earlier stage of fast water current stage. However, with the sediment suspension, the water transparency declined, resulting in rapid attenuation of underwater light intensity, which restrained algae growth. Therefore, the algal biomass at the later period decreased. The concentrations of organic nitrogen and phosphorus ascended steadily with nutrients release. Because of algae uptake, the variations of ammonium and inorganic phosphorus concentrations showed the adverse trend of algal biomass change, which declined at the earlier stage and ascended at the later period during fast water current stage.

5.

Conclusions

In this study, the algal dynamic experiment was conducted in an ecological flume to discover the mechanism of algae growth. The experimental results show that nutrients release resulting from sediment suspension driven by hydrodynamic forces promotes algae growth. Based on the results of ecological flume experiment, an algae growth model is developed. The quantitative relations

between nutrients release and flow velocity are incorporated in the model. By calibration and verification, the quantitative relations between physico-chemical factors and algae growth are determined. Parameters of the algal growth model fit well with literature values, which prove the validity of model. Simulation of algae and nutrients kinetic processes during experiment shows that calculated values reasonably agree with experimental data, with the largest relative error of 16.3%. In conclusion, the developed algal growth model reflects the internal mechanism and dynamics of algal growth, and has good application potential for forecasting future algal blooms in Taihu Lake.

Acknowledgement We would like to acknowledge Nanjing Institute of Geography and Limnology, Chinese Academy of Sciences for the assistance with the ecological flume experiment.

references

Boqiang, Q., Chengxin, F., 2002. Exploration of conceptual model of nutrient release from inner source in large shallow lake. China Environ. Sci. 22, 150–153. Chengxin, F., Boqiang, Q., Yue, S., 1998. Substance exchange across water–sediment interface in Meiliang Bay and Wuli Lake. Lake Sci. 1, 73–78. Colleen, J.H., David, M.C., Mohammad, O., 2005. Simulating nutrient cycling and removal through treatment wetlands in arid/semiarid environments. Ecol. Eng. 1, 25–39. Eran, F., Marcelo, J., Gedalia, S., 2003. Simulation model of wastewater stabilization reservoirs. Ecol. Eng. 2, 121–145. Hongping, P., Jianyi, M., Hong, Z., 2000. The dynamic model of algae in West Lake of Hangzhou. J. Aquat. Creature 2, 144–149. Jørgensen, S.E., 1994. Fundamentals of Ecological Modeling, 2nd ed. Elsevier Science B.V., Netherlands. Leah, B., Edward, P.G., Stephen, G.N., Kevin, F., 2005. Diversity and biomass dynamics of marine algae in Biosphere II’s tropical reef macrocosm. Ecol. Eng. 4, 442–456. Qiujin, X., 2001. Ecological simulation of algae growth in Taihu Lake. Lake Sci. 2, 149–157. Richard, D.B., et al., 2005. Use of a simulation model to examine effects of nutrient loading and grazing on Potamogeton perfoliatus L. communities in microcosms. Ecol. Model. 2–4, 483–512. Sandra, L.M., Renato, W., Jorge, M., 2002. Simulation of alternative management strategies for red algae, luga roja, (Gigartina skottsbergii Setchell and Gardner) in southern Chile. Ecol. Model. 1–2, 121–133. Sebnem, A., Ilgi, K.K., 2006. Batch kinetics of nitrogen and phosphorus removal from synthetic wastewater by algae. Ecol. Eng. 1, 64–70. Shuncai, S., Yiping, H., 1993. Lake Taihu. Coast Press, Beijing. Xiangfei, H., 1999. Analysis and Ecological Investigation of Lake. Standards Press, Beijing. Yiping, L., Yong, P., Kesen, C., Jinping, P., Ling, D., 2004. Study on the starting principles of sediment by water force in Taihu Lake. Adv. Water Sci. 6, 770–774. Yong, P., 1998. General Numerical Study on Atmosphere–Water Environment in Taihu Lake Area. Meteorological Press, Beijing. Yuanbo, L., 1998. Ecological simulation of blue algae bloom in Meiliang bay, Taihu Lake and analysis of its harness strategies. Lake Sci. 4, 53–59.