Quantitative studies on sediment release fluxes and their impact on water quality are important for water pollution control, ecological restoration, water safety, and human health. In this study, we conducted high-frequency, synchronous field observations of meteorology, hydrology, and water quality to determine the relationship between sediment release rate and wind speed in the central region of Lake Taihu. We combined these results with our previous findings from other regions to establish the temporal–spatial variation in sediment release patterns for this lake. We then calculated the annual total nitrogen (TN) and total phosphorus (TP) release fluxes. We constructed an environmental fluid dynamics code (EFDC) model; we then loaded the temporal–spatial variation parameters and simulated the effects on different TN and TP concentrations. Overall, the following results were observed: (1) the critical wind speed at which sediment was first suspended in the central region of the lake was 4 m/s, and the fitted curve of the sediment release rate and wind speed was r = 144.7x−100 (R2 = 0.851); and (2) the annual TN and TP release fluxes of Lake Taihu were approximately 3,086 and 740 tons, respectively. This research would provide a basis for decision-making regarding pollution control in this region.

INTRODUCTION

Lake Taihu is one of China's five major freshwater lakes and one of the three most polluted (Wang & Pei 2013; Zhu et al. 2013). In 2007, a toxic algal bloom occurred that made headlines globally; consequently, pollution control in Lake Taihu has attracted large amounts of interest (Panagopoulos et al. 2007; Xie et al. 2014). Excessive nutrient loading, especially nitrogen (N) and phosphorus (P), has led to water quality deterioration, eutrophication, and large harmful algal blooms (Abell et al. 2010; Xu et al. 2010; Zhang et al. 2014). Numerous studies have revealed that N and P are the most important limiting nutrients for eutrophication (Xu et al. 2010; Chen et al. 2011). The suspension of sediments is considered an important process that impedes the recovery of lakes from eutrophication (Palm et al. 2004), especially when external nutrient sources are controlled (Søndergaard et al. 2003; Ahlgren et al. 2011; Steinberg 2011).

Lake Taihu is a typical wind-driven shallow lake. Wind significantly affects hydrodynamic intensity and results in the suspension of sediments at the surface (Maxam & Webber 2010). This dynamic endogenous release significantly impacts water quality (Qin et al. 2004; Sun et al. 2006) and has been well documented (Gleizon et al. 2003; Kersten et al. 2005; Zhu et al. 2005). Frequent dynamic action in this lake causes up to tens of centimetres of surface sediments to become suspended; as a result, nutrients are released from sediment pore water. After this dynamic effect disappears, the suspended solids (SS) settle and become buried in the sediment; this allows degradation to occur in a reducing environment and sediment to accumulate until the next wind process. A previous study considered the hydrodynamic shear stress as a starting point, and calculated annual total nitrogen (TN) and total phosphorus (TP) release fluxes of 10,570 and 899 t/a, respectively, in a flume experiment (Qin et al. 2006). The annual TP release flux has been estimated at approximately 430 t/a based on a calculation of suspended particulate matter resulting from disturbances under different wind speeds (Fan et al. 2004).

Previous studies have generally focused on suspension and nutrient release flux estimations under this assumption, ignoring the environmental effects resulting from the settlement process. The mathematical models established usually only consider sediment as a constant source term when estimating sediment release fluxes. Furthermore, they do not systematically consider sediment variations across different regions or changes in the dynamic release parameters with wind intensity in conjunction with a water quality model (Li 2005; Hu et al. 2006; Zhu et al. 2011). In this study, we conducted field observations to establish the relationship between wind speed and sediment release in the central region of Lake Taihu. Combined with our previous studies in other regions of the lake (Yan 2008; Hu 2012; Wang et al. 2014, 2015), we calculated the temporal–spatial variation sediment release parameters, and applied these to an environmental fluid dynamics code (EFDC) model, taking into consideration the settling coefficient, to determine incremental TN and TP concentrations in Lake Taihu caused by sediment release.

METHODS

Research area

Lake Taihu (Figure 1) is a large shallow lake with a surface area of 2,338 km2 and an average depth of 1.9 m. The average annual water temperature is 17.3 °C and ranges between 4.8 °C and 29.2 °C. The average annual rainfall in the Taihu Basin is 1,181 mm, with obvious inter-annual variations. The flood season occurs between May and September, and results in more than 60% of the annual rainfall. Generally, the wind speed varies between 0 and 10 m/s, with an average speed of 4.3 m/s in the warm season (southeast) and 0.9 m/s in the cold season (northwest) (Qian 2012). The spatial distributions of TN and TP are heterogeneous, with the highest concentrations recorded in the north, and the lowest in the east. Across the entire lake, the TN and TP concentrations vary between 0.5 and 7 mg/L and 0.02 and 0.35 mg/L, respectively (Deng et al. 2008). The sediment in Lake Taihu is mainly distributed in bulk or in strips. Approximately 1,250 km2 (54%), 970 km2 (42%), and 530 km2 (23%) of the lake has a sediment thickness of 20, 30, and 40 cm, respectively. Silt is mainly distributed in Zhushan Bay, Meiliang Bay, and on the west coast (regions I and II), and in the eastern area (region III).
Figure 1

Sediment survey and experimental points in Lake Taihu.

Figure 1

Sediment survey and experimental points in Lake Taihu.

Field experiment

To effectively study the relationships between wind, hydrodynamic structure, sediment suspension, and nutrient release, we set up an observation platform (Figure 2) in the central area of Lake Taihu (Figure 1; point A: 31°13′44.64″N, 120°6′30.49″E). High-frequency, synchronized, and continuous monitoring was conducted between July 22 and 30, 2014. The monitored variables included wind speed, wind direction, three-dimensional (3D) velocity, and SS concentration.
Figure 2

Diagram of the field observation instrument layout.

Figure 2

Diagram of the field observation instrument layout.

A handheld weather station (PH-II, XPH Company, China) was placed 5 m above the water surface to record the wind speed and direction every 5 min. An acoustic Doppler current profiler (ADP; SonTek Company, Ohio USA) was used to monitor the vertical velocity of the water body and the layer 80 cm above the lake bottom using a sampling frequency of 10 min. The 3D velocity of the surface (50 cm below the water surface) and the bottom layer (5 cm above the water–sediment interface) was monitored using two acoustic Doppler velocimeters (ADV; SonTek Company, Ohio, USA) with an operating frequency of 10 Hz. The water turbidity was continuously recorded using two optical backscatter point sensors (OBS; Campbell Scientific Company, Utah, USA) located at the same height as the two ADV.

Through these observations, we obtained the quantitative relationship between wind speed and sediment suspension rate in the central region of the lake. Following this, based on the effects of wind on the sediment and settlement characteristics of the suspended particles, we estimated sediment suspension and settlement fluxes separately. The net sediment release load equaled the difference between the suspension and the settlement flux. The following formula was used: 
formula
1
where P is the sediment net suspension flux, t; Psus is the sediment suspension flux, t; Pset is the sediment settlement flux, t; S is the sediment coverage, m2; Mi is the sediment suspension rate, g/(m2·d); Nj is the average settlement rate of the sediment, g/(m2·d); and Ti is the wind duration, d.
The sediment suspension rate was calculated as follows: 
formula
2
where is the volume of the water sample in a suspended particulate trap, L; Cn is the nutrient concentration in the water at the nth sampling time, mg/L; C0 is the initial nutrient concentration, mg/L; Vi is the sample volume, L; Cj–1 is the nutrient concentration in the water at the (j–1)th sampling time, mg/L; Ca is the nutrient concentration after adding original water, mg/L; t is the release time, d; and A is the area of the water–sediment interface, m2.
The sediment settlement rate was calculated as follows: 
formula
3
where h is the water depth, m; t is the settling time, d; and C is the concentration after settlement, mg/L.

Numerical simulation

EFDC is an open-source 3D model with an embedded turbulence closure submodel; it is often used to simulate estuarine and coastal ocean circulation hydrodynamics and eutrophication situations. We applied temporal–spatial variation parameters to the submodel to determine the incremental TN and TP concentrations in Lake Taihu caused by sediment release. The governing mass-balance equation for each of the water quality state variables was expressed as follows: 
formula
4
where C is the concentration of a water quality state variable; u, v, and w are velocity components in the x-, y-, and z-directions, respectively; , , and are turbulent diffusivities in the x-, y-, and z-directions, respectively; and represents the internal and external sources and sinks per unit volume.

To better simulate lake topography, the vertical thickness was isometrically divided into three layers based on water depth. Lake Taihu is a typical wind-driven current shallow lake. Atmospheric and surface winds comprised the main dynamic boundary conditions in this model. Daily rainfall data were used to select the average monitoring values obtained from eight weather stations around Lake Taihu. Annual values were utilized from the wind farm data. The model adopted steady initial and boundary conditions with a calculation time of 1 year and a time step of 5 seconds. In this model, the role of sediment diagenesis was not take into account; sediment was only deemed to be a nutrition source and sink. These TN and TP parameters were coupled into the ‘benthic flux of nutrients’ submodule of the EFDC. Positive values were defined as the source, while negative values the sink. The nutrient release rate, which was converted into an hourly flux per unit area (g/hour, with the same frequency as the wind speed), was calculated according to the experimental results. In this research, we only consider the distribution of TN and TP concentrations caused by sediment release (the transformation process between different forms is not considered). So, in the ‘benthic flux of nutrients’ submodule, the given value of ‘benthic flux rate of phosphorus’ is TP rather than PO4-P. The given value of ‘benthic flux rate of ammonia nitrogen and NO2-NO3 nitrogen’ is TN and not just NH4-N and NO2-NO3 (the TN release rate parameter was allocated by the proportion NH4-N:NO2-NO3, between 1:4 and 1:6).

RESULTS AND DISCUSSION

Sediment suspension in the central lake region

Flow field characteristics

During field observations period, Lake Taihu was at the high water level stage, with an average depth of 2.7 m. The weather was predominantly cloudy and the temperature was between 24 °C and 37 °C. Humidity varied between 50% and 90%, while sea level pressure was 990–1,010 MPa. There were occasional rain showers on July 26 and 27.

Figure 3 shows the corresponding relationship between the wind farm and the vertical structure of the instantaneous water flow. The hourly average wind speed varied between 0.5 and 6.1 m/s, while the maximum instantaneous wind speed reached 9 m/s, almost exceeding the normal wind speed range. The corresponding flow velocity varied between 0 and 30 cm/s. Overall, the response relationship between flow and wind was not obvious even at the surface water layer. However, based on the structure of the instantaneous water flow, the flow direction was markedly consistent, and the flow velocity substantially decreased with increasing depth. This was because the wind dramatically changed during the observation period. Although the wind speed was relatively high, since the wind direction constantly changed, the wind fetch was not sufficient to form stable flow farm conditions.
Figure 3

Corresponding relationship between the wind farm and the vertical structure of the instantaneous water flow.

Figure 3

Corresponding relationship between the wind farm and the vertical structure of the instantaneous water flow.

Turbidity distribution

Turbidity at the surface (approximately 50 cm below the water surface) and bottom layer (approximately 5 cm above the water–sediment interface) was continuously recorded using an OBS. Figure 4 shows a positive correlation between wind speed and turbidity. During observations, the turbidity increased with wind speed, regardless of depth (at the water surface or the bottom). The turbidity of the bottom water was always greater than that at the surface at the same time. The turbidity substantially varied between 0 and 50 NTU when the wind speed was less than 4 m/s, and significantly increased when the wind speed increased to 7 m/s, with a greater margin at the bottom than at the surface. Maximum turbidities occurred simultaneously throughout the water body and occurred slightly later than the maximum wind speed. This indicated that it takes time for turbidity to respond to wind. Furthermore, when the wind speed suddenly dropped, the bottom turbidity decreased and then remained constant before slowly decreasing further (Figure 5, after 12:00 on July 25). The increases in turbidity and the layering phenomenon indicated that 4 m/s was the approximate critical wind speed at which the sediment became suspended.
Figure 4

Wind speed and turbidity variations during the observation period.

Figure 4

Wind speed and turbidity variations during the observation period.

Figure 5

SS concentration variations during the monitoring period.

Figure 5

SS concentration variations during the monitoring period.

Sediment release rate

We collected water samples from the bottom (approximately 5 cm above the water–sediment interface), middle (approximately 130 cm below the water surface), and surface (approximately 20 cm below the water surface) of the water body every 3 hours between 08:00 and 16:00 h. During sampling the hourly wind speeds were approximately 1–7 m/s. Figure 5 shows SS concentration variations; values generally remained between 20 and 40 mg/L with the maximum value exceeding 100 mg/L. The SS trend was relatively clear and consistent between both layers. We calculated variations in the sediment suspension rate with wind speed according to the monitored wind speed and corresponding SS concentrations during the experiment (Figure 6). The results were expressed as r = 144.7X−100 (R2 = 0.851; r: release rate, g/(m2·d); X: wind speed, m/s). Wind varied between approximately 0 and 8 m/s during sampling, and the continuously high frequency turbidity monitoring values indicated that less sediment was suspended at wind speeds under 4 m/s. Therefore, the sediment suspension fitting function was suitable for wind speeds of 4–8 m/s. A better fit could not be obtained because the TN and TP concentrations were disturbed by numerous factors during this experiment. Therefore, TN and TP release fluxes were obtained by multiplying the SS fluxes with the corresponding percentages in the sediment.
Figure 6

Curve fitting of SS suspension rate and wind speed.

Figure 6

Curve fitting of SS suspension rate and wind speed.

Sediment release flux estimate and simulation

Sediment suspension and settlement rate in each region

Based on the above analysis, we obtained the sediment release rate for the central lake region of Lake Taihu. According to the sediment distribution survey throughout Lake Taihu (Figure 1), the sediments in Zhushan Bay, Meiliang Bay, and north of the lake center were the most polluted; these findings corresponded with water quality. The west coast was slightly better, and the sediments in the central area and on the south coast had the lowest pollution levels. The sediment distribution results obtained from this survey were comparable with those presented in previous studies (Wang et al. 2012). Combined with the sediment characteristics from our previous studies, we divided Lake Taihu into four sections (Figure 1). Our previous experiments show the SS suspension rate could be expressed as r = 20.72X2.034 in Meiliang Bay (region I; Wang et al. 2014, 2015), r = 314.91X−879.78 on the west coast (region II; Li 2005; Hu 2012), and r = 119.48X−185.1 in the eastern area (region III; Li 2005; Hu 2012). The following quantitative relationship existed between the average sediment settlement flux and wind speed: y = 111.7e0.2186X (R2 = 0.679; y: settlement flux, g/[m2·d]) (Yan 2008; Hu 2012).

Temporal–spatial variations in sediment release parameters

In this model, the wind farm data comprised a year of measured values. The hourly wind speed data are shown in Figure 7. The annual average wind speed was approximately 4.3 m/s, with a maximum value of almost 16 m/s. We sampled 35 sediment points in Lake Taihu (Figure 1) and determined the TN and TP concentrations for each sample. The final nutrient contents in the sediments of each region (Table 1) were averaged from each sampling point in the region.
Table 1

Sediment nutrient concentrations in each region

Regions II III IV 
TN (mg/kg) 1,276 883 900 891 
TP (mg/kg) 382 384 278 314 
Regions II III IV 
TN (mg/kg) 1,276 883 900 891 
TP (mg/kg) 382 384 278 314 
Figure 7

Wind farm conditions in the model (hourly data).

Figure 7

Wind farm conditions in the model (hourly data).

In each region, the fitting functions of the sediment settlement and suspension rate, and the wind speed were used to calculate sediment specific release values in different regions under simulated wind speed conditions. Since sediment suspension did not occur when the wind speed was low, only the settlement process was considered when the wind speed was less than a critical value for a particular region. TN and TP concentrations are affected by many factors; therefore, in this model, the TN and TP specific release values were obtained by multiplying the SS suspension values with the corresponding percentages of TN and TP in the sediment.

Hydrodynamics and water quality calibration

Since the main driving force of this model was wind speed, we focused on wind parameters and on the calibration of the bottom roughness coefficient. The calibration parameters included a roughness coefficient of 0.02 m, a wind drag coefficient of 0.003, and a block coefficient of 1. The daily water levels recorded by four hydrological stations in Lake Taihu were used to calibrate the hydrodynamic parameters (Figure 8). The results of the measured and simulated values showed good agreement. In addition, the average absolute deviations were as follows: 0.1 m (point (1), 2.95%, root-mean-square error (RMSE): 0.117), 0.12 m (point (2), 3.62%, RMSE: 0.148), 0.11 m (point (3), 3.6%, RMSE: 0.137), and 0.07 m (point (4), 2.33%, RMSE: 0.093) (Figure 9).
Figure 8

Water level ((1)–(4)) and water quality (1–30) calibration points.

Figure 8

Water level ((1)–(4)) and water quality (1–30) calibration points.

Figure 9

Water level calibration results.

Figure 9

Water level calibration results.

The changing situation of water quality is very complicated due to the influence of numerous factors on nutrient dynamics (transformation and transport). Therefore, the simulated error relative error (RE) was relatively large (Table 2) (TN: 22%–51%, NH4+: 28%–76%, TP: 24%–56%, Chlorophyll-a [Chl-a]: 10%–52%). In contrast, the simulation results of the central lake and eastern coastal region were relatively better (RE < 30%). Chl-a is one of the initial reaction variables of water eutrophication. In this study, the spatial variation of Chl-a was large. Overall, the water environment mathematical model can better simulate the spatial and temporal changes of the main water quality indexes in Lake Taihu. This provides the foundation for the simulation of the distribution of TN and TP concentration caused by sediment suspension.

Table 2

Water quality calibration results

Regions (points) TN (mg/L)
 
NH4+ (mg/L)
 
TP (mg/L)
 
Chl-a (μg/L)
 
Obs Cal RE Obs Cal RE Obs Cal RE Obs Cal RE 
Meiliang Bay (1–5) 3.6 3.6 32 1.1 0.7 47 0.11 0.09 24 41.9 35.8 32 
Zhushan Bay (6–7) 5.7 4.6 23 3.0 0.9 70 0.16 0.13 24 88.1 53.5 40 
Western coast (8–9) 4.8 3.7 33 1.3 0.8 63 0.14 0.08 52 43.0 46.8 10 
Southern coast (10–14) 2.5 2.4 36 0.2 0.2 70 0.08 0.07 56 22.2 21.0 13 
Eastern lake (15–17) 1.1 1.5 41 0.2 0.1 76 0.04 0.04 32 10.9 13.2 30 
Eastern coast (18–21) 1.7 1.9 29 0.2 0.2 28 0.05 0.06 31 17.7 19.4 12 
Gonghu (22–25) 2.5 2.5 55 0.4 0.2 53 0.08 0.09 33 23.8 14.8 52 
Central lake (26–30) 2.2 2.0 22 0.2 0.4 28 0.08 0.08 28 29.4 20.4 31 
Regions (points) TN (mg/L)
 
NH4+ (mg/L)
 
TP (mg/L)
 
Chl-a (μg/L)
 
Obs Cal RE Obs Cal RE Obs Cal RE Obs Cal RE 
Meiliang Bay (1–5) 3.6 3.6 32 1.1 0.7 47 0.11 0.09 24 41.9 35.8 32 
Zhushan Bay (6–7) 5.7 4.6 23 3.0 0.9 70 0.16 0.13 24 88.1 53.5 40 
Western coast (8–9) 4.8 3.7 33 1.3 0.8 63 0.14 0.08 52 43.0 46.8 10 
Southern coast (10–14) 2.5 2.4 36 0.2 0.2 70 0.08 0.07 56 22.2 21.0 13 
Eastern lake (15–17) 1.1 1.5 41 0.2 0.1 76 0.04 0.04 32 10.9 13.2 30 
Eastern coast (18–21) 1.7 1.9 29 0.2 0.2 28 0.05 0.06 31 17.7 19.4 12 
Gonghu (22–25) 2.5 2.5 55 0.4 0.2 53 0.08 0.09 33 23.8 14.8 52 
Central lake (26–30) 2.2 2.0 22 0.2 0.4 28 0.08 0.08 28 29.4 20.4 31 

Sediment release fluxes and impacts on water quality

The temporal–spatial variation sediment release parameters under simulated wind conditions were loaded into the EFDC model to obtain hydrodynamic results. Therefore, the immediate impacts of the TN and TP concentrations resulting from wind-driven sediment suspension could be simulated. Furthermore, using the hourly temporal–spatial variation parameters and the statistical areas of the four partitions, we obtained the SS suspension fluxes by summing each. Following this, the TN and TP release fluxes were calculated based on the nutrient content test results of the sediment. Since the release of N and P from suspended sediments is a very complex situation, ongoing chemical and biological reactions occur during this process. In addition, the actual wind field was neither long-lasting nor steady, and the sediment in the water body gradually settled after a large storm. During this process, a large amount of insoluble and hard release nutrients settled with the SS to form sediment. Therefore, the net sediment release load was calculated using the suspension flux subtracted from the settlement flux. The annual net TN and TP release flux could be determined using the sediment nutrient concentrations in each region.

The findings revealed that the net sediment release load was 3.1 million tons, and the TN and TP flux was 3,086 and 740 tons, respectively. These values were consistent with the results of Fan et al. (2004) and Hu (2012) who also used TN and TP concentrations in their calculations. N and P pollution in Lake Taihu is mainly derived from the surrounding rivers, atmospheric dry and wet deposition, and internal release. The internal release flux of TN that was calculated in this study accounted for almost 15% of the total TN source of the whole lake, while that of TP accounted for nearly 25%. Therefore, internal pollution comprises a major source of pollution in Lake Taihu. Internal pollution governance such as dredging, especially in Meiliang Bay and the Zhushan Bay, should be included in integrated pollution remediation works.

In the EFDC model, the hourly TN and TP release rate of each region was coupled through ‘the benthic flux of nutrients’ submodule. The prevailing wind direction of Lake Taihu was southeast in summer and northwest in winter. Therefore, we simulated TN and TP concentration increments (only sediment release without exogenous input) under these two conditions (Figure 10). The maximum incremental concentrations of TN and TP under southeast wind conditions could reach 1.8 and 0.13 mg/L, and were slightly lower (1.6 and 0.1 mg/L) under northwest wind conditions. Under southeast wind conditions, the highly concentrated areas were mainly distributed on the northwest side of the two bays, and under northwest wind conditions these were mainly distributed on the northeast side. This was because these two bays are semi-enclosed and located in the northern part of Lake Taihu. Under the prevailing wind condition, circumfluence can form in the bay, consequently the released nutrients are not effectively spread, resulting in poor water quality within the two bays.
Figure 10

Distributions of the maximum instantaneous concentrations of TN and TP under two typical wind conditions.

Figure 10

Distributions of the maximum instantaneous concentrations of TN and TP under two typical wind conditions.

CONCLUSIONS

In the central area of Lake Taihu, the critical wind speed required to cause sediment suspension is 4 m/s. The turbidity at this wind speed only slightly changed, and water stratification was not significant. When the wind speed reached 7 m/s, the turbidity at the bottom rapidly increased over a short period, indicating that the structure of the surface sediments had been destroyed and that the suspended sediment concentration was high. Based on synchronous meteorological, hydrodynamic, and water quality monitoring data, we observed that the SS concentration was positively correlated with wind speed. The functional relationship between the sediment suspension rate and wind speed could be expressed as r = 144.7X−100 (R2 = 0.851). Compared with the northern lake and western coastal regions, the suspension flux at the same wind speeds was much lower. This difference was mainly associated with sediment distribution and characteristics. Northwest Lake Taihu is located in a heavily polluted region. The background nutrient concentration of the water body is relatively high, and under the influence of the prevailing wind and lake flow, cyanobacteria are predominantly distributed in the northwest coastal region. Algae generally become enriched and die, then deposit to the lake bottom to form a flowing mud layer. Relatively small intensity disturbances can re-suspend sediment. Furthermore, flow, bottom shear stress, and sediment suspension are influenced by aquatic vegetation. The density and the distribution of the vegetation influences the flow velocity to varying degrees, and flow resistance increases with plant density. The bottom shear stress is lower in a vegetated area compared with a non-vegetated area. The presence of the vegetation helps to resist deformation and erosion of the sediment bed, maintains bed stability, and improves water quality by removing suspended particles. According to SS suspension rate, hourly wind data, and sediment nutrient concentration, we calculated the annual TN and TP release fluxes of the entire lake to be approximately 3,086 and 740 t. The simulation results of the EFDC model show that nutrients cannot diffuse well in Meiliang Bay and Zhushan Bay, and significant endogenous pollution mainly occurs along the western coast and in the northern regions. These findings should be considered for sediment dredging and eutrophication treatment in Lake Taihu.

ACKNOWLEDGEMENTS

This work was financially supported by the National Natural Science Foundation of China (51609116 and 41371222), Natural Science Foundation of Jiangsu province (BK20160961) and the Startup Foundation for Introducing Talent of NUIST (2016r21). We would like to thank Hohai University for the experimental platform, and all of the people who assisted in our research processes.

REFERENCES

REFERENCES
Ahlgren
J.
Reitzel
K.
De Brabandere
H.
Gogoll
A.
Rydin
E.
2011
Release of organic P forms from lake sediments
.
Water Res.
45
(
2
),
565
572
.
Chen
J.
Quan
W. T.
Sun
J. H.
2011
Relationship between nitrogen, phosphorus concentration and water quality factors
.
Environmental Monitoring of China
27
(
3
),
79
83
(in Chinese)
.
Deng
J. C.
Chen
Q.
Zhai
S. J.
Yang
X. C.
Han
H. J.
Hu
W. P.
2008
Spatial distribution characteristics and environmental effect of N and P in water body of Taihu Lake
.
Environ. Sci.
29
(
12
),
3382
3386
(in Chinese)
.
Fan
C. X.
Zhang
L.
Qin
B. Q.
Wang
S. M.
Hu
W. P.
Zhang
C.
2004
Estimation on dynamic release of phosphorus from wind-induced suspended particulate matter in Lake Taihu
.
Science in China: Series D: Earth Sciences
47
(
8
),
710
719
.
Hu
K. M.
2012
Study on Emissions Reducing Distribution of Pollutants Based on Water Quality Targets Achievement and Traceability Technology in Lake Taihu
.
PhD Thesis
,
Hohai University
,
Nanjing, China
(in Chinese)
.
Hu
C. H.
Hu
W. P.
Zhang
F. B.
Hu
Z. X.
Li
X. H.
Chen
Y. G.
2006
Sediment resuspension in the Lake Taihu, China
.
Chin. Sci. Bull.
51
,
731
737
.
Li
Y. P.
2005
Study on the Influence Factors of Transparency of Experiment and Numerical Model in Taihu Lake
.
PhD Thesis
,
Hohai University
,
Nanjing, China
(in Chinese)
.
Palm
A.
Cousins
I.
Gustafsson
Ö.
Axelman
J.
Grunder
K.
Broman
D.
Brorström-Lunden
E.
2004
Evaluation of sequentially-coupled POP fluxes estimated from simultaneous measurements in multiple compartments of an air–water–sediment system
.
Environ. Pollut.
128
(
1–2
),
85
97
.
Qian
H. Z.
2012
The Influence of Wind Field to the Spatial Distribution of Chlorophyll-a Concentration
.
PhD Thesis
,
Nanjing University of Information Science and Technology
,
Nanjing, China
(in Chinese)
.
Qin
B. Q.
Zhu
G. W.
Zhang
L.
Luo
L. C.
Gao
G.
Gu
B. H.
2006
Estimation of internal nutrient release in large shallow Lake Taihu, China
.
Science in China: Series D: Earth Sciences
49
(
Suppl. 1
),
38
50
.
Søndergaard
M.
Jensen
J. P.
Jeppesen
E.
2003
Role of sediment and internal loading of phosphorus in shallow lakes
.
Hydrobiologia
506
,
135
145
.
Steinberg
C. E. W.
2011
Aerobic phosphorus release from shallow lakes
.
Sci. Total Environ.
409
(
21
),
4640
4641
.
Sun
X. J.
Zhu
G. W.
Luo
L. C.
Qin
B. Q.
2006
Experimental study on phosphorus release from sediments of shallow lake in wave flume
.
Science in China: Series D: Earth Sciences
49
(
Suppl. 1
),
92
101
.
Wang
P.
Lu
S. Y.
Wang
D. W.
Xu
M. S.
Gan
S.
Jin
X. C.
2012
Nitrogen, phosphorus and organic matter spatial distribution characteristics and their pollution status evaluation of sediments nutrients in lakeside zones of Taihu Lake
.
China Environmental Science
32
(
4
),
703
709
(in Chinese)
.
Wang
J. J.
Pang
Y.
Li
Y. P.
Huang
Y. W.
Jia
J. J.
Zhang
P.
Kou
X. P.
2014
The regularity of wind-induced sediment resuspension in Meiliang Bay of Lake Taihu
.
Water Sci. Technol.
70
(
1
),
167
174
.
Wang
J. J.
Pang
Y.
Li
Y. P.
Huang
Y. W.
Luo
J.
2015
Experimental study of wind-induced sediment suspension and nutrient release in Meiliang Bay of Lake Taihu, China
.
Environ. Sci. Pollut. R.
22
(
14
),
10471
10479
.
Xu
H.
Paerl
H. W.
Qin
B. Q.
Zhu
G. W.
Gao
G.
2010
Nitrogen and phosphorus inputs control phytoplankton growth in eutrophic Lake Taihu, China
.
Limnol. Oceanog.
55
(
1
),
420
432
.
Yan
R. R.
2008
Study on the Influence Factors of Water Bloom and Pollution Load in Lake Taihu
.
PhD Thesis
,
Hohai University
,
Nanjing, China
(in Chinese)
.
Zhu
M. Y.
Zhu
G. W.
Wang
Y. P.
2011
Influence of scum of algal bloom on the release of N and P from sediments of Lake Taihu
.
Environ. Sci.
32
,
409
416
(in Chinese)
.
Zhu
M. Y.
Zhu
G. W.
Zhao
L. L.
Yao
X.
Zhang
Y. L.
Gao
G.
Qin
B. Q.
2013
Influence of algal bloom degradation on nutrient release at the sediment–water interface in Lake Taihu, China
.
Environ. Sci. Pollut. R.
20
,
1803
1811
.