Simulation of the Parameters Effecting the Water Quality Evolution of Xuanwu Lake, China

International Journal of Environmental Research and Public Health

Article Simulation of the Parameters Effecting the Water Quality Evolution of Xuanwu Lake, China

Min Pang 1, Weiwei Song 2,* , Yuan Liu 3 and Yong Pang 2,*

1 School of Environmental Science and Engineering, Southern University of Science and Technology, Shenzhen 518055, China; [email protected] 2 Key Laboratory of Integrated Regulation and Resource Development on Shallow Lakes, Ministry of Education, College of Environment, Hohai University, Nanjing 210098, China 3 Jiangsu Environmental Engineering Technology Company Limited, Nanjing 210036, China; [email protected] * Correspondence: [email protected] (W.S.); [email protected] (Y.P.)

Abstract: After years of water environment improvement, China’s water quality has improved to some extent in recent years. However, different water areas have different characteristics of water pollution. The paper used mathematical models to investigate the influence of different parameters on the water quality of Xuanwu Lake, China. The predominant focus was on the nutrients concentration due to changing the amount of pollutants, degradation coefficient, water diversion discharge and diffusion coefficient. The results showed that the amount of pollutants had the most significant impact, followed by the degradation coefficient. The total phosphorus and total nitrogen concentrations of Xuanwu Lake increased with the increase of the amount of pollutants. The water quality of Xuanwu Lake decreased significantly with the increase of degradation coefficient. Increasing the water diversion discharge will not only make a big difference in water quality, but it will also worsen the water quality. The effect of the amount of pollutants on Xuanwu Lake total phosphorus and total Citation: Pang, M.; Song, W.; Liu, Y.; nitrogen is 4.1 and 5.7 times that of water diversion discharge. The influence of total phosphorus and Pang, Y. Simulation of the Parameters total nitrogen in the degradation coefficient scheme is 3.5 and 6.2 times that of the water diversion Effecting the Water Quality Evolution of Xuanwu Lake, China. Int. J. discharge scheme. The diffusion coefficient has almost no effect on the water quality of Xuanwu Lake. Environ. Res. Public Health 2021, 18, From the practical difficulty and implementation effect of water environment improvement, the order 5757. https://doi.org/10.3390/ of water quality improvement effect from good to bad is as follows: the amount of pollutants scheme, ijerph18115757 degradation coefficient scheme, water diversion scheme, diffusion coefficient scheme. Under the circumstance of limited water diversion, the lake will effectively improve the water quality. Reducing Academic Editor: Hyun-Sang Shin the discharge of pollutants is the fundamental measure to control water environment problems, and water diversion is an auxiliary measure to improve the water ecology. It will become a trend to Received: 7 April 2021 combine the reduction of pollutant discharge and water transfer for water environment improvement. Accepted: 25 May 2021 This paper is of significance for improving the water quality of Xuanwu Lake, and it also provides a Published: 27 May 2021 scientific method for water environment improvement of water diversion projects.

Publisher’s Note: MDPI stays neutral Keywords: water quality improvement; inﬂuence parameters; model simulation; Xuanwu Lake with regard to jurisdictional claims in published maps and institutional afﬁl- iations.

1. Introduction Water is not only the origin of life, but also promotes human progress and social development. The comprehensive utilization of water resources has promoted economic Copyright: © 2021 by the authors. development. Comprehensive utilization of water resources can be divided into non-water- Licensee MDPI, Basel, Switzerland. This article is an open access article consuming river use and water-consuming outside of river use. The water utilization in distributed under the terms and non-water-consuming river use includes hydropower, water transportation, ﬁsheries, water conditions of the Creative Commons entertainment, and other ecosystem services. The water utilization in water-consuming Attribution (CC BY) license (https:// outside of river use includes agricultural, industrial and domestic. With the intensiﬁed creativecommons.org/licenses/by/ development of water resources by mankind, rivers have been severely damaged, including 4.0/). the water environment [1,2]. Pollution of the water environment has led to the destruction

Int. J. Environ. Res. Public Health 2021, 18, 5757. https://doi.org/10.3390/ijerph18115757 https://www.mdpi.com/journal/ijerph Int. J. Environ. Res. Public Health 2021, 18, 5757 2 of 16

of the ecological status of water, causing water bodies to lose their functions, which in turn has restricted human development. After the water environment is damaged, the ecological function of water bodies is restored by reducing pollutants discharge and water diversion [3,4]. By constructing the urban sewage pipe network, sewage enters the plant for treatment, thereby reducing direct discharge into river [5,6]. Construction of a new sewage pipe network is in conflict with the old urban infrastructure, and will also take many years to complete [7,8]. In addition, there are certain problems in the operation and management of the sewage pipe network, which makes it impossible to use reasonably, making it useless [9,10]. It is difficult to control discharge of pollutants into the river. Generally, water with better quality is diverted to avoid polluted areas through ecological water regulation to meet water quality standards [11,12]. In recent years, water diversion has gradually become a commonly used pollution control method as it can improve water quality in short time. Water-diversion projects include the construction of manmade structure that modify the natural flow of a waterway. Diversion projects are usually developed for purposes of flood control, hydroelectric power generation, water supply, and farm irrigation. Clean freshwater is a valuable resource and is in short supply worldwide. Water diversion engineering has been successfully applied and implemented with various suitable conditions, including water diversion engineering, artificially diverting water from an existing sur- plus to a watershed or river to redistribute much-needed water supplies, and changing living conditions and the ecological environment [13]. There are more than 160 large-scale diversion projects built in 24 countries [14]. Lake Okeshobe reduces total phosphorous concentration by diverting the course of the Kissimmee River in the United States. In China, water diversion projects have increased in recent years. The South to North Water Diversion Project is a famous water diversion project, with a total length of 3833 km [15]. It is regarded as a strategic and ambitious solution to resolve water shortage problems in the north with three routes delivering water from the Yangtze River to the north of China [16]. The famous Great Canal from Beijing to Hangzhou with a total length of 1794 km, is the longest and first manual canal in the world [17]. The water diversion project has also been implemented in more places and has played a key role in water environment protection [18–28]. Mathematical models are useful scientific tools and technical means for studying water environment. The mathematical model is based on the conservation of mass, conservation of energy, and conservation of momentum, simulating the physical, chemical, and biological phenomena in water. Model calculation cannot only simulate the experiments that physical models cannot carry out, but also has the advantage of high-cost performance. Mathematical models of water environment are often used in the formulation of water resources protection and water environment treatment plans. The model is used to study the hydrodynamic characteristics, pollutants migration and transformation characteristics, so as to provide a scientific basis and theoretical support for water environment management. The water environment mathematical model consists of hydrodynamic models and water quality models. In the mid-to-late 1970s, mathematical models were used to quantitatively simulate migration, transformation, and changes in water quality. Since the 1980s, researchers have developed CREAMS (chemical runoff and erosion from agricultural management systems), ANSWERS (the areal non-point source watershed environment response simulation), and so on. In this stage, the water quality model research focused on application and gradually improved, forming the QUAL, MIKE, WASP, SWAT, OTIS, BASINS, AGNPSl, ANSWERS and EFDC model systems. The MIKE series model developed by the Danish Institute of Hydraulics is one of the most widely used commercial models, covering Mike Urban, Mike21, Mike3, Litpack, Mike Animator, Mike C-map, Eeco Lab, Mike11, Mike Flood, Feflow, Mike She, Mike Basin, and Mike21C, etc. In recent years, despite the increase in the amount of water diversion upstream of Xuanwu Lake, its water quality has not improved. In this paper, the mathematical model of the water environment is used to study the influence of the main parameters on the water Int. J. Environ. Res. Public Health 2021, 18, 5757 3 of 16

environment of Xuanwu Lake. By collecting topographic, water quality, hydrodynamic and meteorological data of Xuanwu Lake, we construct a two-dimensional mathematical model of Xuanwu Lake water environment. We obtain pollution source information through on- site investigation and sampling. Subsequently, we simulate the current water quality and use the monitoring data to verify the model so that the model can make accurate predictions. According to the current situation survey, the main factors affecting Xuanwu Lake’s water quality are as follows: the amount of pollutants (W), the degradation coefﬁcient (K), the amount of water diversion (Q) and the diffusion coefﬁcient (D). To investigate the effect of these four parameters on the water environment of Xuanwu Lake, we studied each parameter separately while keeping other parameters unchanged. Considering that a large quantity of outer water cannot enhance the water quality of Xuanwu Lake, we propose self-circulation measures of the lake’s water body to study whether it can improve the lake’s water quality.

2. Study Area and Methods 2.1. Study Area Xuanwu Lake is a typical shallow urban lake with an aquatic area of 3.7 square kilometers and drainage area of 5.5 square kilometers. Where the surface water level is 10 m, the mean water depth is 1.14 m and the storage capacity is 4,290,000 m3. The highest water level is 11.15 m (Figure1, SE), and the smallest possible surface water level is 9.8 m, annual surface water level 9.8 m to 10.2 m. The Xuanwu Lake area has a humid and warm subtropical climate. The average annual temperature is 15 ◦C to 16 ◦C. Typically, the warmest temperature is between July and August, with an average of around 28 ◦C. The average annual rain is around 1000 mm. The average annual wind velocity is 8 m/s, and the prevailing wind direction is southeast and northwest [4]. Since 2000, Xuanwu Lake has implemented water diversion engineering, and the water supply reached 50,000 m3/d. In 2003, the water supply reached to 180,000 m3/d, and the next year, it became 280,000 m3/d. The water supply for diversion comes from the Yangtze River, which is directly discharged into Xuanwu Lake through underground concealed pipes after sedimentation. There is almost no suspension of sediment. After the water is diverted through two water plants, it enters Xuanwu Lake from different locations through two pipelines, of which the No. 2 pipeline is divided into six points. The lake’s water then passes through four gates to supply water to the city’s inland rivers. The pollution sources of the lake are mainly ﬁve tributaries into the lake. Xuanwu Lake is divided into four lake areas, namely Northeast (NE) Lake, Southeast (SE) Lake, Southwest (SW) Lake and Northwest (NW) Lake. Under normal circumstances, the water supply of No. 1 water plant is 80,000 m3/d, and it enters Xuanwu Lake through Entrance A. The water supply of No. 2 water plant is 200,000 m3/d, and it enters Xuanwu Lake through the inlet BCDEF under a pipeline. The water supply to the lake through No. 3 tributary is 70,000 m3/d. The total amount of water transferred from Xuanwu Lake is 280,000 m3/d. Therefore, the total water transfer capacity of Xuanwu Lake is 350,000 m3/d (Figure1). According to the water quality evaluation results of Xuanwu Lake in the past three years, the main pollutants exceeding the standard of water quality are total phosphorus and total nitrogen. Under normal water transfer conditions, the water quality concentration of Xuanwu Lake cannot meet the standard (GB 3838-2002) Class IV (TP less 0.1 mg/L, TN less 1.5 mg/L). Int. J. Environ. Res. Public Health 2021, 18, 5757 4 of 16 Int. J. Environ. Res. Public Health 2021, 18, x FOR PEER REVIEW 4 of 18

1 2 China

a B Study location A ① 118.80°E,32.08°N

C N 300m NW 1,2 Water Plant b ② Water quality monitoring point ③ A,B,C,D,E,F Water diversion inlet D a,b,c,d Water outlet

① ② ③④ ⑤ Sewage discharge inlet ④ SW Land

SE E ⑤

c F

d FigureFigure 1. Study 1. Study area. area.

2.2.2.2. Study Study Methods Methods 2.2.1.2.2.1. Hydrodynamic Hydrodynamic Two-Dimensional Two-Dimensional Model Model The two-dimensional hydrodynamic governing equations are the continuous equations The two-dimensional hydrodynamic governing equations are the continuous equa- and momentum equations integrated along the depth direction of the three-dimensional tions and momentum equations integrated along the depth direction of the three-dimen- Reynolds Navier–Stokes average equation of incompressible fluids in the Cartesian coordinate sional Reynolds Navier–Stokes average equation of incompressible fluids in the Cartesian system [29,30]. The equations are as follows: coordinate system [29,30]. The equations are as follows: Z η Z η ℎ𝑢= 𝑢𝑑𝑧,hu =ℎ𝑣̅ =udz 𝑣𝑑𝑧, hv, = vdz, (1) −d −d (1)

where:where:(us ,(𝑢vs),𝑣represents) represents the the rate rate of dischargeof discharge into into the the water water body; body;ρ0 represents𝜌 represents the the densitydensity of water. of water. ContinuityContinuity equation: equation: ∂h ∂hu ∂hv 𝜕ℎ 𝜕ℎ𝑢 𝜕ℎ𝑣+̅ + = hQ (2) + +∂t =ℎ𝑄∂x ∂y (2) 𝜕𝑡 𝜕𝑥 𝜕𝑦 Momentum equation: Momentum equation: 2 2 ∂hu ∂hu ∂hvu ∂η h ∂Pa gh ∂ρ τsx τbx 1 ∂Sxx ∂Sxy ∂ ∂ + 𝜕ℎ𝑢+ 𝜕ℎ𝑢 = f𝜕ℎ𝑣𝑢 hv− gh − 𝜕𝜂 −ℎ 𝜕𝑃 +𝑔ℎ 𝜕𝜌− 𝜏− 𝜏 1 +𝜕𝑆 𝜕𝑆+ (hT𝜕 xx) + hT𝜕 xy + husQ (3) ∂t ∂x + ∂y + = 𝑓∂ℎ𝑣x ̅ −𝑔ℎρ0 ∂x − 2ρ0 ∂x− ρ0 +ρ0 − ρ0 −∂x ∂y+ +∂x (ℎ𝑇)∂+y ℎ𝑇+ℎ𝑢𝑄 (3) 𝜕𝑡 𝜕𝑥 𝜕𝑦 𝜕𝑥 𝜌 𝜕𝑥 2𝜌 𝜕𝑥 𝜌 𝜌 𝜌 𝜕𝑥 𝜕𝑦 𝜕𝑥 𝜕𝑦

2 2 ∂hv ∂hv𝜕ℎ𝑣̅ ∂h𝜕ℎ𝑣vu ̅ 𝜕ℎ𝑣𝑢 ∂η h ∂Pa gh ∂ρ τsy τby 1 ∂Syx ∂Syy ∂ ∂ + + + =+− f hu − gh − − + − − + + hTxy + hTyy + hvsQ (4) ∂t ∂y 𝜕𝑡 ∂x𝜕𝑦 𝜕𝑥 ∂y ρ0 ∂y 2ρ0 ∂y ρ0 ρ0 ρ0 ∂x ∂y ∂x ∂y (4) 𝜕𝜂 ℎ 𝜕𝑃 𝑔ℎ 𝜕𝜌 𝜏 𝜏 1 𝜕𝑆 𝜕𝑆 𝜕 𝜕 =−where:𝑓ℎ𝑢−𝑔ℎx, y are− Cartesian− coordinates;+ −t is time;− h is total+ water+ depth;ℎ𝑇η+is water ℎ𝑇 level;+ℎ𝑣u and𝑄 v 𝜕𝑦 𝜌 𝜕𝑦 2𝜌 𝜕𝑦 𝜌 𝜌 𝜌 𝜕𝑥 𝜕𝑦 𝜕𝑥 𝜕𝑦 are average water depth; ρ is water density; f = 2Ωsinϕ represents the Coriolis factor (ϕ is the geographicalwhere: x, latitude y are CartesianΩ, is the angular coordinates; velocity t ofis thetime; Earth’s h is total rotation); waterSxx depth;, Sxy and η isSyy waterrepresent level; 𝑢 and 𝑣̅ are average water depth; ρ is water density; f = 2Ωsinφ represents the Coriolis factor (φ is the geographical latitude Ω, is the angular velocity of the Earth’s rotation); 𝑆,

Int. J. Environ. Res. Public Health 2021, 18, x FOR PEER REVIEW 5 of 18

𝑆 and 𝑆 represent the radiation stress tensors; Q represents the point source emissions; g represents the gravitational acceleration; and Pa represents the atmospheric pres- Int. J. Environ. Res.sure Public. Health 2021, 18, 5757 5 of 16 Transverse stress Tij, involves viscous resistance, dynamic advection friction resistance, and turbulent frictional resistance, which can be calculated using the vortical viscosity equation for average vertical velocity: the radiation stress tensors; Q represents the point source emissions; g represents the gravitational 𝜕𝑢 𝜕𝑢 𝜕𝑣̅ 𝜕𝑣̅ acceleration; and Pa represents the atmospheric pressure. 𝑇 =2𝐴 ,𝑇 = 𝐴 + ,𝑇 =2𝐴 . (5) 𝜕𝑥Transverse stress𝜕𝑦 T𝜕𝑥ij, involves viscous𝜕𝑥 resistance, dynamic advection friction resistance, and turbulent frictional resistance, which can be calculated using the vortical viscosity equation for average vertical velocity: 2.2.2. Two-Dimensional Water Quality Model ∂u ∂u ∂v ∂v The water quality equations are basedTxx = 2onA the, Tmassxy = balanceA + equation., Tyy = The2A model. that (5) meets the calculation purpose must be two-dimensional∂x or three-dimensional,∂y ∂x ∂whilex considering that the data is suitable for a two-dimensional model. The results of two-dimen- 2.2.2. Two-Dimensional Water Quality Model sional calibration can reflect the actual situation of hydrodynamic and water quality The water quality equations are based on the mass balance equation. The model that meets [31,32]. The two-dimensional water quality transport equation is as follows: the calculation purpose must be two-dimensional or three-dimensional, while considering that the 𝜕𝐶 𝜕𝐶 data𝜕𝐶 is suitable𝜕 for𝜕𝐶 a two-dimensional𝜕 𝜕𝐶 model. The results of two-dimensional calibration can reflect +𝑈 +𝑉 = 𝐸 + 𝐸 +𝐾𝐶 +𝑆 , (6) 𝜕𝑡 𝜕𝑥 the𝜕𝑦 actual𝜕𝑥 situation 𝜕𝑥 of hydrodynamic𝜕𝑦 𝜕𝑦 and water quality [31,32]. The two-dimensional water quality transport equation is as follows: where: Ci represents the water pollutants concentration; Ex and Ey are the diffusion coefficients in the x and y axes, respectively;∂Ci ∂ uC,i v are∂ Cthei flow∂ velocity∂Ci components∂ ∂Ci in the x and + U + V = Ex + Ey + KiCi + Si, (6) y axes, respectively; Ki is the pollutant∂t ∂degradationx ∂y coefficient;∂x ∂x and∂y Si represents∂y the pollutant sediment release coefficient. where: Ci represents the water pollutants concentration; Ex and Ey are the diffusion coefficients in the x and y axes, respectively; u, v are the flow velocity components in the x and y axes, respectively; Ki 2.3. Model Setupis and the Calibration pollutant degradation coefficient; and Si represents the pollutant sediment release coefficient. 2.3.1. Model Setup 2.3. Model Setup and Calibration The Xuanwu Lake model is subdivided into a mixed network of three quadrants of 2.3.1. Model Setup angles with mesh spacing from about 20 to 30 m [11,17]. Assuming that the lake surface is The Xuanwu Lake model is subdivided into a mixed network of three quadrants of angles stationary at the initial moment, there is no disturbance, and the time step t = 1 d [29,30]. with mesh spacing from about 20 to 30 m [11,17]. Assuming that the lake surface is stationary at the The grid and topographyinitial moment, of the there Xuanwu is no disturbance, Lake model and are the shown time step in Figuret = 1 d [2.29 The,30]. depth The grid at and topography the floor of theof bottom the Xuanwu of Xuanwu Lake model Lake are ranges shown from in Figure 0.8 m2. Theto 2.2 depth m, atthe the deepest floor of area the bottom is of Xuanwu Southwest Lake,Lake and ranges the deepest from 0.8 water m to 2.2depth m, the is 2.2 deepest m [33,34]. area is According Southwest Lake,to the and measured the deepest water depth topography of isthe 2.2 wading m [33,34 area]. According of Xuanwu to the Lake measured and the topography bottom elevation of the wading of the area model, of Xuanwu Lake and through modelthe verification, bottom elevation the lake of’s the Manning model, throughcoefficient model is 38 verification, m1/3/s, and the the lake’s eddy Manning cur- coefficient is rent viscosity is38 0.28 m1/3 [35,36]./s, and the eddy current viscosity is 0.28 [35,36].

(a) (b)

Figure 2. (a) XuanwuFigure Lake 2. model(a) Xuanwu grid. (b Lake) Bottom model elevation. grid. (b ) Bottom elevation.

2.3.2. Boundary Conditions Hydrodynamic parameters: the temperature is 10 ◦C, the initial water level is 10 m, and the ﬂow rate is 0 at the start [9,10,37]. The water level of the gate at the outlets a, b, c, d is 3 m. The temperature, wind velocity, rainfall and average monthly inﬂow rate for 2017 are shown in Figure3. The meteorological data source comes from website (http://data.sheshiyuanyi.com/WeatherData/, accessed on February 2021) and water plant. Int.Int. J.J. Environ.Environ. Res.Res. PublicPublic HealthHealth 2021,, 18,, xx FORFOR PEERPEER REVIEWREVIEW 6 of 18

2.3.2. Boundary Conditions Hydrodynamic parameters: the temperature is 10 °C, the initial water level is 10 m, and the flow rate is 0 at the start [9,10,37]. The water level of the gate at the outlets a, b, c, d is 3 m. The temperature, wind velocity, rainfall and average monthly inflow rate for Int. J. Environ. Res. Public Health 2021, 18, 5757 6 of 16 2017 are shown in Figure 3. The meteorological data source comes from website (http://data.sheshiyuanyi.com/WeatherData/, accessed on February 2021) and water plant.

Temperature Rainfall Wind speed No.No. 11 waterwater plantplant 4040 Temperature Rainfall 3030 Wind speed 3.53.5 /s) No. 2 water plant 3 250 /s) No. 2 water plant

250 3 3535 3.0 2525 3.0 30 200 30 200 2.5 20 2.5 2525 20 150 2.0 20 150 2.0 20 1515 100 1.5 1515 100 1.5 10

10 Rainfall (mm) 10 1.0 10 Rainfall (mm)

Temperature (℃) Temperature 50 1.0 Wind speed(km/h)

Temperature (℃) Temperature 50

5 Wind speed(km/h) 5 55 0.5 0 0.5 00 0 0 Total inflow discharge(m 0.0 0 Total inflow discharge(m 0.0 2017-1-12017-1-1 2017-5-1 2017-5-1 2017-9-1 2017-9-1 2017-1-12017-1-1 2017-5-1 2017-5-1 2017-9-1 2017-9-1 2017-1-12017-1-1 2017-5-1 2017-5-1 2017-9-1 2017-9-1 2017-32017-3 2017-6 2017-6 2017-9 2017-9 2017-12 2017-12 Date Date Date Date Date Date Date Date FigureFigure 3. 3. DailyDaily temperature, temperature, rainfall, rainfall, wind wind speed, speed, and and monthl monthlyy total total inflow inﬂow discharge discharge of of Xuanwu Xuanwu Lake Lake in in 2017. 2017.

WaterWater qualityquality parameters: parameters: the the pollutant pollutant (TP, TN) (TP, concentration TN) concentration of the inﬂow of the and inflow of the boundary and of thetheofthe boundaryboundary a, b, c, and ofof thethe d outlets a,a, b,b, c,c,are andand as dd shown outletsoutlets in areare Figure asas shownshown4. Water inin quality FigureFigure data4.4. WaterWater were qualityquality obtained datadata from werewere the obtainedNanjing Environmental from the Nanjing Protection Environmental Bureau. Protection Bureau.

Inlet boundary TN a 0.160.16 InletInlet boundaryboundary TPTP 2.42.4 Inlet boundary TN 0.250.25 aa a 2.82.8 b bb b 0.14 2.2 2.6 0.14 2.2 0.20 c 2.6 cc 0.20 c 2.4 0.12 d 2.4 dd 0.12 2.02.0 d 2.2 0.15 2.2 0.10 0.15 2.0 0.10 1.8 2.0 1.8 1.8 0.08 0.10 1.8 0.08 0.10 1.6 1.61.6 1.6 0.060.06 1.41.4 1.4 0.050.05 Inlet boundary TP (mg/L)TP Inlet boundary Inlet boundary TN (mg/L) boundaryInlet TN 1.4 1.2 Outlet boundary (mg/L) TP Outlet Inlet boundary TP (mg/L)TP Inlet boundary Outlet boundary TN (mg/L) Outlet boundary TN Inlet boundary TN (mg/L) boundaryInlet TN 1.2 Outlet boundary (mg/L) TP Outlet 0.040.04 (mg/L) Outlet boundary TN 2017-32017-3 2017-6 2017-6 2017-9 2017-9 2017-12 2017-12 2017-32017-3 2017-6 2017-6 2017-9 2017-9 2017-12 2017-12 2017-32017-3 2017-6 2017-6 2017-9 2017-9 2017-12 2017-12 2017-32017-3 2017-6 2017-6 2017-9 2017-9 2017 2017-- Date Date Date Date Date Date Date Date FigureFigure 4. 4. InletInletInlet andand and outletoutlet outlet boundaryboundary boundary waterwater water ququ qualityality of of Xuanwu Xuanwu Lake Lake in in 2017. 2017. 2.3.3. Model Calibration 2.3.3. Model Calibration This paper uses a trial-and-error method to calibrate and verify the model parameters. Accord- This paper uses a trial-and-error method to calibrate and verify the model parameters. Accordinging to the calculation to the calculation of the 2D of water the quality2D water model quality [38– 40model], the TN[38–40], degradation the TN coefficient degradation K of AccordingXuanwu Lake to the was calculation 8.55 × 10−7 ofs− 1theto 1.412D water× 10− 6qualitys−1, and model the TP [38–40], degradation the TN coefficient degradation K was coefficient K of Xuanwu Lake was 8.55 × 10−−77 s−−11 to 1.41 × 10−−66 s−−11, and the TP degradation coefficient5.92 × 10−7 Ks− of1 to2.26 Xuanwu× 10 −Lake6 s−1 was. The 8.55 pollutants × 10 (TP,s to TN) 1.41 concentration × 10 s , and calibration the TP and degradation verification −7 −1 −6 −1 coefficientin the districts K was of NE, 5.92 SE, × 10 NW,−7 ss− SW1 to2.26to2.26 are shown ×× 1010−6 s ins−1.. Figure TheThe pollutantspollutants5. Table1 shows (TP,(TP, TN)TN) that concentrationconcentration the error between cal-cal- Int. J. Environ. Res. Public Health 2021, 18, x FOR PEER REVIEW 7 of 18 ibrationibrationmeasured andand and verificationverification calculated data inin thth ofe waterdistricts pollutants of NE, inSE, the NW, whole SW lake are of shown different in monthsFigure were5. Table less 1than shows 20%. that It can the be error seen thatbetween the model measured can be usedand forcalculatedlculated model simulation datadata ofof waterwater and prediction pollutantspollutants [41 –inin43 thethe]. whole lake of different months were less than 20%.20%. ItIt cancan bebe seenseen thatthat thethe modelmodel cancan bebe 0.35 0.28 Calculatedused for model simulation Calculated and prediction0.24 [41–43]. Calculated Calculated Measured 0.30 Measured Measured 0.24 Measured 0.24 0.20 0.25 0.20 0.20 0.16 0.20 0.16 0.16 0.12 0.12 0.12 0.15 0.10 0.08 0.08 SE-TP (mg/L)

NE-TP (mg/L) NE-TP 0.08 SW-TP (mg/L) NW-TP (mg/L) 0.04 0.05 0.04 0.04 0.00 0.00 0.00 0.00 2016-11-19 2017-4-19 2017-9-19 2016-11-19 2017-4-19 2017-9-19 2016-11-19 2017-4-19 2017-9-19 2016-11-19 2017-4-19 2017-9-19 Date Date Date Date 4.0 3.2 Calculated Calculated 2.8 Calculated 2.8 Calculated Measured 3.5 Measured Measured Measured 2.8 2.4 2.4 3.0 2.4 2.0 2.0 2.5 2.0 2.0 1.6 1.6 1.6 1.2 1.2 1.2 1.5

SE-TN (mg/L)SE-TN 0.8 0.8 NE-TN (mg/L) NE-TN SW-TN (mg/L) SW-TN 0.8 1.0 NW-TN (mg/L) 0.5 0.4 0.4 0.4 0.0 0.0 0.0 0.0 2016-11-19 2017-4-19 2017-9-19 2016-11-19 2017-4-19 2017-9-19 2016-11-19 2017-4-19 2017-9-19 2016-11-19 2017-4-19 2017-9-19 Date Date Date Date Figure 5. Model (water quality) validationvalidation of Xuanwu Lake in 2017.

Table 1. Error analysis table.

TP TN Time Simulation Measured Simulation Measured K (s−1) Error (%) K (s−1) Error (%) (mg/L) (mg/L) (mg/L) (mg/L) 2017-1-3 1.53 × 10−6 0.0811 0.0775 4.65 1.1 × 10−6 1.7385 1.7875 2.74 2017-2-7 2.26 × 10−6 0.0608 0.0525 15.81 9.26 × 10−7 2.1463 2.125 1.00 2017-3-1 1.53 × 10−6 0.0793 0.0775 2.39 9.74 × 10−7 2.1613 1.9825 9.02 2017-4-6 1.35 × 10−6 0.0929 0.0875 6.14 1.27 × 10−6 1.5486 1.5525 0.25 2017-5-4 1.01 × 10−6 0.1269 0.1175 7.99 1.29 × 10−6 1.5589 1.52 2.56 2017-6-2 6.77 × 10−7 0.1713 0.175 2.10 1.41 × 10−6 1.5040 1.3825 8.79 2017-7-3 1.44 × 10−6 0.0988 0.0825 19.70 1.24 × 10−6 1.5623 1.5875 1.58 2017-8-1 5.92 × 10−7 0.1805 0.2 9.74 1.47 × 10−6 1.5150 1.3125 15.43 2017-9-4 6.58 × 10−7 0.1980 0.18 10.01 1.13 × 10−6 1.7373 1.7475 0.58 2017-10-2 1.21 × 10−6 0.1125 0.0975 15.39 8.55 × 10−7 2.3060 2.3075 0.07 2017-11-1 8.61 × 10−7 0.1141 0.1375 17.05 1.1 × 10−6 1.9879 1.7325 14.74 2017-12-4 1.25 × 10−6 0.1113 0.095 17.20 1.1 × 10−6 1.9409 1.7575 10.44

2.4. Calculation Schemes To study the relationship between water quality and the influencing factors of Xuanwu Lake, we set different schemes for the amount of pollutants (W), degradation coefficient (K), water diversion (Q) and diffusion coefficient (D). Based on the current amount of pollutants and reduction plan, 11 schemes were set up to calculate the amount of pollutants after reduction [7,44], and the relationship between the amount of pollutants and the water quality of Xuanwu Lake was studied (Table S1). Based on the current range of degradation coefficients, 27 schemes were set up to calculate the water quality under different degradation coefficients [45], and the relationship between water pollutants concentration and degradation coefficients of Xuanwu Lake was shown in Table S2. Based on the water diversion capacity of 350,000 m3/d and current actual amount of diversion, 16 types of water diversion schemes were calculated, and the relationship between the water pollutant concentration and water diversion of Xuanwu Lake was studied (Table S3). We also studied 19 kinds of diffusion coefficient schemes (Table S4). The schemes and results are presented in Table S4. We calculated the total phosphorus (TP) and total nitrogen (TN)

Int. J. Environ. Res. Public Health 2021, 18, 5757 7 of 16

Table 1. Error analysis table.

2.4. Calculation Schemes To study the relationship between water quality and the influencing factors of Xuanwu Lake, we set different schemes for the amount of pollutants (W), degradation coefficient (K), water diversion (Q) and diffusion coefficient (D). Based on the current amount of pollutants and reduction plan, 11 schemes were set up to calculate the amount of pollutants after reduction [7,44], and the relationship between the amount of pollutants and the water quality of Xuanwu Lake was studied (Table S1). Based on the current range of degradation coefficients, 27 schemes were set up to calculate the water quality under different degradation coefficients [45], and the relationship between water pollutants concentration and degradation coefficients of Xuanwu Lake was shown in Table S2. Based on the water diversion capacity of 350,000 m3/d and current actual amount of diversion, 16 types of water diversion schemes were calculated, and the relationship between the water pollutant concentration and water diversion of Xuanwu Lake was studied (Table S3). We also studied 19 kinds of diffusion coefficient schemes (Table S4). The schemes and results are presented in Table S4. We calculated the total phosphorus (TP) and total nitrogen (TN) of the southeast (SE), northeast (NE), southwest (SW), northwest (NW), and the whole lake (Whole).

3. Results and Discussion 3.1. Impact of Pollutant Quantity (W) on Water Quality Keeping other conditions unchanged, the water quality change of Xuanwu Lake is calculated by reducing the emission concentration of pollutants. It can be seen from Figure6a that the TP and TN concentrations of the lake increase with the increase in the amount of pollutants. Due to the order of magnitude difference in the concentration of TN and TP, the average concentration of nutrients is dominated by TN. According to routine monitoring of water quality data, the average concentration of water pollutants of the Northeast Lake is the highest of the four lake regions, and the average water quality of the Southwest Lake is the lowest. The pollutant concentration in the Southeast Lake area is close to that of the whole lake. The overall order of water quality from good to bad is Southwest Lake, Northwest Lake, Southeast Lake, Northeast Lake. As shown in Figures1 and6a,b, the main reason why the pollutants concentration of the western lake area is better than the eastern lake area is that pollutants are mainly discharged from the eastern half of the lake. The maximum pollutant emission plan is the status quo. The concentration of TP is 0.0884 mg/L (less than the standard value of 0.1 mg/L), and the water quality of TN is 1.53 mg/L (more than the standard value of 1.5 mg/L). As long as there is one item that does not meet the standard, the conclusion is that it does not meet the standard overall. After reducing TN pollutants by 6.7%, TN was just able to meet the standard. When there are no pollutant discharged into Xuanwu Lake, and only non-point source pollution exists, the TP and the TN concentration are 0.03 mg/Land 0.81 mg/L, respectively. Int. J. Environ. Res. Public Health 2021, 18, 5757 8 of 16 Int. J. Environ. Res. Public Health 2021, 18, x FOR PEER REVIEW 9 of 18

0.10 SE SE 1.6 0.09 NE NE SW SW 0.08 NW NW 1.4 Whole Whole 0.07 1.2 0.06

0.05 1.0 TP concentration (mg/L) concentration TP 0.04 (mg/L) concentration TN 0.8 0.03

0.02 0.6 0 100 200 300 400 0 1000 2000 3000 4000 5000 6000 7000 W(TP) (ton/year) W(TN) (ton/year) (a) A1 A3 A5 A7 A9

A1 A3 A5 A7 A9

(b)

FigureFigure 6.6. ((aa)) WaterWater qualityquality trendtrend ofof variousvarious lakelake areasareas inin XuanwuXuanwu LakeLake withwith thethe dischargedischarge ofof pollutants.pollutants. ((bb)) TheThe firstfirst rowrow shows the TP concentration (mg/L) fields from left to right for Schemes A1, A3, A5, A7, and A9; the second row shows the shows the TP concentration (mg/L) fields from left to right for Schemes A1, A3, A5, A7, and A9; the second row shows the TN concentration (mg/L) fields from left to right for Schemes A1, A3, A5, A7, and A9. TN concentration (mg/L) fields from left to right for Schemes A1, A3, A5, A7, and A9.

3.2. ImpactBased onof Degradation the data analysis Coefficient results, we(K) calculatedon Water theQuality concentration fields of scheme A1, A3, A5, A7, andWe A9 calculate (Table S1), the respectively, reasonable as showndegradation in Figure coefficient6b. The red [4] area schemes in the figure of Xuanwu represents Lake the areawhile where keeping the water other quality conditions is not unchanged. up to standard. Within It can the be degradation seen that as the coefficient amount of of pollutants Xuanwu decreases,Lake, the thechange high-concentration range of TP was of water 0.036~0.099 quality mg/L, gradually and decreases the change and range the low-concentration of TN was 0.62 area gradually increases. Although the amount of TP and TN pollutants is reduced by the same to 1.72 mg/L, as shown in Figure 7a. On the whole, the water pollutants concentration of proportion, the concentration of TP can meet the standard more easily than TN. Xuanwu Lake decreased significantly with the increase in the degradation coefficient. The 3.2.concentration Impact of Degradation trend of water Coefficient quality (K) in on all Water lake areas Quality is the same, and the rate of concen- trationWe decline calculate gradually the reasonable decreases. degradation However, coefficient Southwest [4] schemes Lake of with Xuanwu the best Lake water while keepingquality otherhas the conditions largest unchanged. change trend, Within followed the degradation by Southeast coefficient Lake, of Xuanwu then Lake, Northwest the change Lake range and of TPNortheast was 0.036~0.099 Lake. The mg/L, pollutant and the changeconcentrations range of TN of wasthe 0.62lake to areas 1.72 mg/L, (KTP as< 1 shown × 10−6 in, KFigureTN < 75a ×. On10−7 the), from whole, large the to water small pollutants are Southeast concentration Lake, Northeast of Xuanwu Lake, Lake decreased Northwest significantly Lake, and with South- the increasewest Lake. in the When degradation the degradation coefficient. coefficient The concentration K is less trend than of 0.000008 water quality d−1, the in allwater lake areasquality is theof Northeast same, and theLake rate is of greater concentration than that decline of graduallySoutheast decreases. Lake. The However, lake area Southwest with the Lake largest with thedecline best waterrate of quality pollutant has the concentration largest change has trend, the followed smallest by pollutant Southeast concentration. Lake, then Northwest In general, Lake × −6 × −7 andthe Northeastpollutant Lake. concentration The pollutant of concentrations Southeast Lake of the and lake Northwest areas (KTP < Lake 1 10 are, KessentiallyTN < 5 10 the), from large to small are Southeast Lake, Northeast Lake, Northwest Lake, and Southwest Lake. When same, which is the same as that of the whole lake. That is, the water quality of Southeast the degradation coefficient K is less than 0.000008 d−1, the water quality of Northeast Lake is greater thanLake that and of Northwest Southeast Lake.Lake can The represent lake area with the av theerage largest water decline quality rate ofof pollutantthe whole concentration lake. When hasthe thedegradation smallest pollutant coefficient concentration. is very small, In general, the water the pollutant quality of concentration the four lakes of Southeast is essentially Lake andthe same. Northwest With Lake the increase are essentially of degradation the same, which coefficient, is the same the pollutant as that of theconcentration whole lake. change That is, therate water of Southeast quality of Lake Southeast and LakeNorthwest and Northwest Lake is Lakeessentially can represent the same, the average which water the difference quality of

Int. J. Environ. Res. Public Health 2021, 18, x FOR PEER REVIEW 10 of 18

between the pollutant concentration of the Northeast Lake Area and the Southwest Lake Area becomes larger, and the pollutant concentration difference from other lake areas becomes larger. We calculated the concentration fields for the B27, B13, B7, B4, and B1 schemes (Table Int. J. Environ. Res. Public Health 2021,S2),18, 5757 as shown in Figure 7b. The TP of each scheme can meet water quality standard9 of(GB 16 3838-2002) Class IV, while the TN of B27 and B13 schemes cannot. Red represents areas where the water quality is not up to standard. It can also be seen in Figure 7b that as the degradation coefficient increases, the red area gradually decreases, the blue area gradu- the whole lake. When the degradation coefﬁcient is very small, the water quality of the four lakes is ally increases, and the water quality improves. It can be seen that the water pollutants essentially the same. With the increase of degradation coefﬁcient, the pollutant concentration change concentrationrate of Southeast of Lake each and lake Northwest area changes Lake is asyn essentiallychronously the same, with which the degradation the difference betweencoefficient. the pollutantWith the concentrationincrease in the of thedegradation Northeast Lakecoefficient, Area and the the order Southwest from greatest Lake Area to becomes lowest of larger, the lakeandthe areas pollutant that have concentration improved difference water qualit from othery is Southwest lake areas becomes Lake, Northwest larger. Lake, South- east Lake, and Northeast Lake. 0.12 1.8

0.10 1.6

1.4

0.08 1.2

1.0 0.06 0.8

0.04 SE 0.6 SE NE

TP concentration (mg/L) concentration TP NE TN concentration (mg/L) concentration TN 0.4 SW SW 0.02 NW 0.2 NW Whole Whole 0.0 0.00 0.0 2.0x10-6 4.0x10-6 6.0x10-6 8.0x10-6 1.0x10-5 0.0 2.0x10-6 4.0x10-6 6.0x10-6 8.0x10-6 1.0x10-5 K(TP) K(TN) (a)

B27 B13 B7 B4 B1

(b)

Figure 7.7. ((a)) WaterWater qualityquality trendtrend ofof variousvarious areasareas inin XuanwuXuanwu LakeLake withwith thethe degradationdegradation coefficientcoefficient (K). (b) The firstfirst row shows the TP concentration (mg/L) fields from left to right for Schemes B27, B13, B7, B4, and B1; the second row shows shows the TP concentration (mg/L) fields from left to right for Schemes B27, B13, B7, B4, and B1; the second row shows the the TN concentration (mg/L) fields from left to right for Schemes B27, B13, B7, B4, and B1. TN concentration (mg/L) fields from left to right for Schemes B27, B13, B7, B4, and B1.

3.3. ImpactWe calculated of Discharge the concentration (Q) on Water fields Quality for the B27, B13, B7, B4, and B1 schemes (Table S2), as shownUnder in Figure the7 conditionb. The TP ofthat each the scheme water can diversio meet watern concentration quality standard is kept (GB constant, 3838-2002) reason- Class ableIV, while water the diversion TN of B27 schemes and B13 schemesare calculated cannot. to Red obtain represents the water areas concentration where the water of qualityXuanwu is Lake,not up as to standard.shown in ItFigure can also 8a. be The seen variation in Figure7 rangb thate asof theTP degradationconcentration coefficient in each increases, lake area the is 0.0767red area to gradually 0.0926 mg/L, decreases, and the bluevariation area gradually range of increases,TN is 1.344 and to the 1.592 water mg/L. quality The improves. variation It can be seen that the water pollutants concentration of each lake area changes asynchronously with the degradation coefficient. With the increase in the degradation coefficient, the order from greatest to lowest of the lake areas that have improved water quality is Southwest Lake, Northwest Lake, Southeast Lake, and Northeast Lake.

3.3. Impact of Discharge (Q) on Water Quality Under the condition that the water diversion concentration is kept constant, reasonable water diversion schemes are calculated to obtain the water concentration of Xuanwu Lake, as shown in Int.Int. J. Environ. J. Environ. Res. Res. Public Public Health Health2021 2021, 18, 18, 5757, x FOR PEER REVIEW 11 of10 18 of 16

Figureinterval8a. of The TP variation and TN range concentration of TP concentration of the pollutant in each amount lake area (W) is 0.0767 schemes to 0.0926 is 4.1 mg/L,and 5.7 and thetimes variation that of range the concentration of TN is 1.344 variation to 1.592 mg/L. range The of this variation scheme. interval The variation of TP and interval TN concentration of TP ofand the TN pollutant concentration amount (W)of the schemes degradation is 4.1 and coeffic 5.7 timesient (K) that schemes of the concentration is 3.5 and 6.2 variation times that range ofof this the scheme.concentration The variation variation interval range ofof TP this and scheme. TN concentration With the increase of the degradationin diversion coefficient flow, (K)the schemes pollutant is 3.5concentration and 6.2 times of thateach of lake the concentrationarea increased variation slightly. range The order of this of scheme. growth With rate the increaseof pollutant in diversion concentration flow, the in pollutant each lake concentration area from large of each to lakesmall area is Southeast increased slightly. Lake, North- The order ofwest growth Lake, rate Southwest of pollutant Lake, concentration Northeast Lake. in each In lakethis areascheme, from the large order to smallfrom islarge Southeast to small Lake, Northwestof the overall Lake, water Southwest concentration Lake, Northeast for each Lake.lake is, In Northeast this scheme, Lake, the Southeast order from Lake, large North- to small of thewest overall Lake, water and concentrationSouthwest Lake. for eachHowever, lake is, af Northeastter the diversion Lake, Southeast flow became Lake, Northwestgreater than Lake, 3 and150,000 Southwest m3/d, the Lake. TP However,concentration after in the Southeast diversion La flowke was became greater greater than than the TN 150,000 concentra- m /d, the TPtion concentration in Northeast in Lake. Southeast We compare Lake was this greater scheme than with the TN the concentration pollutant quantity in Northeast scheme Lake. (K). We comparePlan A5 thisand scheme Plan C16 with add the the pollutant same amount quantity of schemeTP and (K).TN Planinto the A5 andriver. Plan We C16 compare add the this same amount of TP and TN into the river. We compare this scheme with the pollutant quantity scheme (K). scheme with the pollutant quantity scheme (K). In the case of 350,000 m3/d water diver- In the case of 350,000 m3/d water diversion, the concentrations of TP and TN are 0.062 mg/L and sion, the concentrations of TP and TN are 0.062 mg/L and 1.1 mg/L, respectively. Without 1.1 mg/L, respectively. Without water diversion, the concentrations of TP and TN are 0.0854 mg/L water diversion, the concentrations of TP and TN are 0.0854 mg/L and 1.47 mg/L, respec- and 1.47 mg/L, respectively. tively.

1.60 0.092

0.090 1.55

0.088 1.50 0.086 SE 0.084 NE SE 1.45 SW NE 0.082 NW SW 1.40 Whole TP concentration (mg/L) TP 0.080 NW concentration (mg/L) TN Whole 0.078 1.35

0.076 -5.0x104 0.0 5.0x104 1.0x105 1.5x105 2.0x105 2.5x105 3.0x105 3.5x105 4.0x105 -5.0x104 0.0 5.0x104 1.0x105 1.5x105 2.0x105 2.5x105 3.0x105 3.5x105 4.0x105 Q (m3/d) Q (m3/d) (a)

C1 C4 C7 C11 C16

(b)

FigureFigure 8. 8.(a ()a Water) Water quality quality trendtrend of various lake lake areas areas in in Xuanwu Xuanwu Lake Lake with with the discharge the discharge (Q). ( (Q).b) The (b )first The row first shows row showsthe TP concentration (mg/L) fields from left to right for Schemes C1, C4, C7, C11, and C16; the second row shows the TN the TP concentration (mg/L) fields from left to right for Schemes C1, C4, C7, C11, and C16; the second row shows the TN concentration (mg/L) fields from left to right for Schemes C1, C4, C7, C11, and C16. concentration (mg/L) fields from left to right for Schemes C1, C4, C7, C11, and C16. We calculated the TP and TN concentration fields for the C1, C4, C7, C11, and C16 We calculated the TP and TN concentration fields for the C1, C4, C7, C11, and C16 schemes schemes (Table S3, Figure 8b). The results reveal that with the increase in water diversion, (Table S3, Figure8b). The results reveal that with the increase in water diversion, the concentration field in each lake area changes little. As with other schemes, the pollutant concentration in the western district of lake is better than the left lake area. It can be seen from the figure that the area

Int. J. Environ. Res. Public Health 2021, 18, x FOR PEER REVIEW 12 of 18

the concentration field in each lake area changes little. As with other schemes, the pollu- Int. J. Environ. Res. Public Health 2021, tant18, 5757 concentration in the western district of lake is better than the left lake area. It can11 ofbe 16 seen from the figure that the area exceeding the TP standard is small, and most of the lake areas can meet the standard. However, the area exceeding the TN standard is very large, and most of the lake areas cannot meet the standard. This phenomenon is consistent with exceedingthe results the of TP 16 standardcalculation is small, schemes. and most of the lake areas can meet the standard. However, the area exceeding the TN standard is very large, and most of the lake areas cannot meet the standard. This3.4. Impact phenomenon of Diffusion is consistent Coefficient with (D) the on results Water of 16Quality calculation schemes. 3.4. ImpactWe calculated of Diffusion the Coefficientwater concentration (D) on Water of QualityXuanwu Lake under different diffusion coefficient schemes while keeping other conditions unchanged. The diffusion coefficient of We calculated the water concentration of Xuanwu Lake under different diffusion coefficient water pollutants is mainly affected by temperature and velocity [4]. When the value of the schemes while keeping other conditions unchanged. The diffusion coefficient of water pollutants is mainlydiffusion affected coefficient by temperature ranges from and velocity0.1 to 9, [4 the]. When pollutants the value concentration of the diffusion of coefficienteach lake rangesarea fromdoes 0.1not to change 9, the pollutants significantly. concentration However, of each with lake the area increase does not in change the diffusion significantly. coefficient, However, withthere the was increase a slight in theincrease diffusion in water coefficient, quality there in wasthe Northwest a slight increase Lake in area, water where quality TP in in- the Northwestcreased by Lake 0.0015 area, mg/L where and TP TN increased increased by by 0.0015 0.02 mg/L mg/L. and Such TN small increased changes by 0.02 are mg/L.negligible Such smallfor current changes pollutant are negligible concentrations. for current We pollutant calculated concentrations. the TP and TN We calculatedconcentration the TPfields and for TN concentrationthe D1, D5, D10, fields D14, for the and D1, D19 D5, schemes D10, D14, (Table and D19 S4), schemes respectively, (Table S4),as shown respectively, in Figure as shown 9b. It in Figurecan be9b. seen It can from be seen the fromfigure the that figure under that underdifferent different diffusion diffusion coefficients, coefficients, the the concentration concentration fieldfield in in each each lake lake area area is almostis almost the the same. same.

1.60 0.092

0.090 1.55

0.088 1.50 0.086 SE 0.084 1.45 NE SE SW NE NW

TP concentration (mg/L) concentration TP 0.082 SW (mg/L) concentration TN Whole NW 1.40 0.080 Whole

0.078 1.35 0246810 0246810 D D (a)

D1 D5 D10 D14 D19

(b)

Figure 9. (a) Water quality trend of various lake areas in Xuanwu Lake with the diffusion coefficient (D). (b) The first row shows the TP concentration (mg/L) fields from left to right for Schemes D1, D5, D10, D14, and D19; the second row shows the TN concentration (mg/L) fields from left to right for Schemes D1, D5, D10, D14, and D19.

3.5. Analysis of Lake Water Body Self-Circulation

According to previous calculations, Southwest Lake has the lowest pollutants concentration and the best water quality, this is due to its low mobility, which makes it is difﬁcult for pollutants to reach Southwest Lake. With the consideration of saving water resources and energy, diversion Int. J. Environ. Res. Public Health 2021, 18, 5757 12 of 16

water from Southwest Lake to Northwest Lake will greatly improve the self-circulation of the lake’s water body. The transfer of water from Southwest Lake to Northwest Lake is conducive to improving the uniformity of water quality, thereby improving water quality. Under certain lake diversion flow (350,000 m3/d), as the self-circulating flow (1 m3/s, 2 m3/s, 3 m3/s, 4 m3/s, 5 m3/s) in the lake increases, the water pollutant concentration in each lake area decreases. However, the pollutants concentration in the East Lake area is still higher than in other lake areas. The water quality intervals are the same. The water quality in the East Lake area is the same as other situations, but the water concentration in the West Lake area decreases significantly with the increase in water flow in the lake. The results show that under the condition of water diversion of 350,000 m3/d, the water quality of the condition of 5 m3/s self-circulation is the best. Without water self-circulation in the lake, the pollutant concentration in the lake tends to increase with the increase of the water flow. However, with the increase in self-circulation flow, water quality has improved significantly, especially in the West Lake area, but water quality of the Northeast Lake did not change with the increase in self-circulation flow. The flow field is virtually unchanged under different self-circulating flows. There is only a slight change in the bridge hole. Generally, the local flow velocity increases with the increase in the amount of water diversion.

3.6. Discussion The two-dimensional mathematical model of water environment used in this paper to study the lake water environment can meet the purpose of calculation entirely and reflect the real situation of hydrodynamic and water quality. The equation discrete method of the model is solved by the explicit finite volume method at the center of the unit, which ensures the conservation of water and momentum in the calculation domain. The model uses unstructured triangle or quadrilateral grid elements, which is more conducive to fitting complex boundary lines. It is convenient and quick to use the wet and dry grid judgment method to deal with the moving boundary of the tidal flat. Use of parallel computing can greatly reduce the computing time. The model is suitable for simulating wide rivers, lakes, and oceans, but it is not suitable for narrow rivers because of its difficulty in dividing the grid. Keeping other conditions unchanged, the TP and TN concentrations of Xuanwu Lake increase with the increased amount of pollutants. The TP concentration range was 0.0268 to 0.092 mg/L, and the TN concentration range was 0.735 to 1.59 mg/L. The overall order of water quality from good to bad is Southwest Lake, Northwest Lake, Southeast Lake, Northeast Lake. Reducing pollutant emissions is the fundamental way to improve water quality. The reduction of pollutants entering Xuanwu Lake is mainly achieved by reducing the direct discharge of domestic sewage from surrounding residents. For areas without a sewage pipe network, the construction of the pipe network should be strengthened to connect sewage to a sewage treatment plant. At the same time, lake silt dredging needs to be carried out regularly due to the impact of sediment on lake water quality in the future. In contrast, the most important factor affecting the water environment of an artificial lake is the amount of pollutants discharged. Controlling the discharge of pollutants controls the quality of the water environment. The pollutants of lakes in nature change very little, and only human factors cause water pollution. Surrounding Xuanwu Lake are urban residents. Domestic sewage is discharged into the lake through the river channel. The interference of humans increases the pollutants in the lake and causes the deterioration of water quality. The Funil Reservoir in Brazil [46] and Lake Okeechobee [47] in the United States are affected by human agriculture, and the water quality changes seasonally. Xuanwu Lake is different from the natural lake Loktak Lake, and is the same as Funil Reservoir and Lake Okeechobee, which are also interfered by humans. In the calculation of the degradation coefficient schemes, the change range of TP was 0.036 to 0.099 mg/L, and the change range of TN was 0.62 to 1.72 mg/L, within the suitable degradation coefficient of Xuanwu Lake. The water quality of Xuanwu Lake decreased significantly with the increase in the degradation coefficient. The pollutants concentrations from large to small lake areas are, Southeast Lake, Northeast Lake, Northwest Lake, and Southwest Lake. The order in value of the decline rate of pollutant concentration is completely opposite to that of the pollutant concentration. The water quality of Southeast Lake and Northwest Lake can represent the water quality of the whole lake, allowing the reduction of the monitoring frequency of lake water quality. If the degradation coefficient is adjusted by human interference, the water quality of Xuanwu Lake can also be improved. The degradation coefficient of water body can be improved by spreading degrading bacteria, rationally adjusting water, widening river channels, and dredging silt. In the calculation of the diffusion coefficient schemes, when the value of the diffusion coefficient ranges from 0.1 to 9, the pollutant concentration of each lake area does not change significantly. However, Int. J. Environ. Res. Public Health 2021, 18, 5757 13 of 16

with the increase in the diffusion coefficient, there was a slight increase in water quality in the Northwest Lake area; TP concentration increased by 0.0015 mg/L and TN concentration increased by 0.02 mg/L. The main factors affecting the diffusion coefficient are temperature and flow rate. In the water quality parameter study of Igapó I Lake (Londrina, Paraná, Brazil) [48], Shagawa Lake (MN, USA) [49], and Koumoundourou Lake (brackish lake, Eleusis Bay), the range of degradation coefficient and diffusion coefficient were also studied. Xuanwu Lake measurements show differences when compared with the degradation coefficients and diffusion coefficients of lakes in different parts of the world, but they all have a range of values. It can be seen from the comparison of the data that the water quality cannot be improved quickly by only increasing the water diversion. From the perspective of improving water quality, reducing pollutants is more effective than water transfer. It is more time-efficient to improve water quality through joint application to reduce some pollutants and water diversion. In the calculation of water diversion schemes, the change range of TP concentration in each lake area is 0.0767 to 0.0926 mg/L, and the change range of TN is 1.344 to 1.592 mg/L. With the increase in water diversion flow, the pollutant concentration of each lake area increased slightly. The number of pollutant schemes affected the TP and TN concentration of Xuanwu Lake by 4.1 and 5.7 times as much as the discharge schemes. By increasing the degradation coefficient, the TP and TN concentrations in Xuanwu Lake were reduced by 3.5 and 6.2 times that of the water diversion schemes. The water quality of Xuanwu Lake cannot be improved by only increasing the diversion water, and doing so would reduce the water quality and cause unnecessary waste of resources. This paper takes the discharge of pollutants during the dry season as an example; combined with the ecological protection needs of Xuanwu Lake, its suitable water diversion flow is 180,000 m3/d. With the self-circulation flow, the water quality of Xuanwu Lake has improved significantly, especially under the condition of 350,000 m3/d. With the increase in self-circulation flow, the water quality of Xuanwu Lake will also improve; however, the flow field in the lake area has not changed significantly thus far. It can be seen that the lake self-circulation can be implemented when the water diversion capacity of the lake is limited or when the water quality cannot be changed with water diversion. Lake self-circulation can not only improve the water quality of the lake, but also avoid damage to the lake’s flow field. Based on the research of this paper, measures to improve the water quality of Xuanwu Lake are proposed. Reducing the discharge of pollutants is the most fundamental measure to improve the quality of the water environment. Improving the quality of the diversion water will also improve the water environment of Xuanwu Lake to a certain extent. It is recommended to open up Southwest Lake and Northwest Lake to make the water flow self-circulate in the lake area, thereby improving water quality. It is recommended to construct a water pump with a volume of one cubic meter in the Southwest Lake, so that the water in the Southwest Lake can be transferred to the Northwest Lake, and the water can be self-circulated, thereby improving the water quality. Based on the real data collected in Xuanwu Lake, the research method of mathematical mod- elling was used to determine the factors influencing changes of water quality parameters. The MIKE21 mathematical model has been widely used all over the world, with theoretical and practical foundations, and reliability. The research results of this paper are also applicable to small lakes with a water exchange cycle of one week, which has certain promotional value. The limitation of this study is that it is not applicable to large shallow lakes, such as Taihu Lake (China) and Lake Okeechobee (USA). Xuanwu Lake is different from large lakes and lakes in nature. Generally, it is small artificial lakes that exist in fast-growing cities. The main issues of small lakes addressed in this paper are the amount of pollutants, degradation coefficient, water diversion, and diffusion coefficient, and does not involve the problem of cyanobacteria blooms and sediment resuspension. This paper has certain reference value for the water environment management of small lakes. With the development of global urbanization and the increase in human demand for water landscapes, there will be more and more small artificial lakes similar to Xuanwu Lake. The water environment of the lake will gradually deteriorate with human intervention. The study found that reducing the discharge of pollutants is the key to improving the water quality of small artificial lakes. In the long term, due to the gradual implementation of the construction of the sewage pipe network, pollutants are gradually discharged into the sewage plant, which gradually improves the lake water quality. Appropriate aeration reoxygenation, artificial hydraulic circulation and diafiltration purification can change the degradation coefficient and diffusion coefficient, and can also improve water quality. Int. J. Environ. Res. Public Health 2021, 18, 5757 14 of 16

4. Conclusions Aimed at the fact that the pollutants concentration of Xuanwu Lake cannot reach the standard stably, we have carried out research on factors influencing water quality with the two-dimensional water environment mathematical model. The effects of pollutants quantity (W), degradation coefficient (K), water diversion (Q), and diffusion coefficient (D) on the water quality of Xuanwu Lake were studied. The purpose was to find out the influence of factors on the water quality of Xuanwu Lake, and to propose water quality improvement measures. The main conclusions are as follows: (1) Keeping other conditions unchanged, the TP and TN concentrations of Xuanwu Lake increase with the increase in the amount of pollutants. The TP concentration range was 0.0268~0.092 mg/L, and the TN concentration range was 0.735~1.59 mg/L. The overall order of water quality from low to high is Southwest Lake District, Northwest Lake District, Southeast Lake District, and Northeast Lake District. Reducing pollutants emissions is the fundamental way to improve water quality. The reduction of pollutants entering Xuanwu Lake is mainly achieved by reducing the direct discharge of domestic sewage from surrounding residents. For areas without a sewage pipe network, the construction of the pipe network should be strengthened so that sewage can be connected to a sewage treatment plant for treatment. At the same time, lake silt dredging needs to be carried out regularly to reduce the impact of sediment on lake water quality. (2) In the calculation of the degradation coefficient scheme, within the degradation coefficient of Xuanwu Lake, the change range of TP was 0.036~0.099 mg/L, and the change range of TN was 0.62~1.72 mg/L. The water quality of Xuanwu Lake decreased significantly with the increase of degradation coefficient. The water quality concentrations from large to small lake areas are Southeast Lake, Northeast Lake, Northwest Lake, and Southwest Lake. The rate of decline of water quality concentration is completely opposite to that of water quality concentration. The water quality of Southeast Lake and Northwest Lake can represent the water quality of the whole lake, which can reduce the monitoring frequency of lake water quality. If the degradation coefficient is adjusted by human interference, the water quality of Xuanwu Lake can also be improved. (3) In the calculation of the water diversion scheme, the variation range of TP concentration in each lake area is 0.0767~0.0926 mg/L, and the variation range of TN is 1.344~1.592 mg/L. With the increase in the diversion flow, the water quality concentration of each lake area increased slightly. The reduction in Xuanwu Lake concentration by reducing TP and TN was 4.1 and 5.7 times that of the water diversion plan. By increasing the degradation coefficient, the TP and TN concentrations in Xuanwu Lake were reduced by 3.5 and 6.2 times that of the water diversion plan. The water quality of Xuanwu Lake cannot be improved by only increasing the replenishment flow, and doing so would reduce the water quality and cause unnecessary waste of resources. This paper takes the discharge of pollutants during the dry season as an example, and combined with the ecological protection needs of Xuanwu Lake, its suitable replenishment flow is 180,000 m3/d. (4) In the calculation of the diffusion coefficient scheme, when the value of the diffusion coefficient ranges from 0.1 to 9, the water quality concentration of each lake area does not change significantly. However, with the increase in the diffusion coefficient, there was a slight increase in water quality in the Northwest Lake area, where TP increased by 0.0015 mg/L and TN increased by 0.02 mg/L. The main factors affecting the diffusion coefficient are temperature and flow rate. The increase in temperature makes the water body diffuse faster, and the temperature has no influence on the water quality. (5) With the self-circulation flow, the water quality of Xuanwu Lake has improved significantly, especially under the condition of 350,000 m3/d. With the increase in self-circulation flow, the water quality of Xuanwu Lake will also improve, but the flow field in the lake area has not changed significantly thus far. It can be seen that the lake self-circulation can be implemented when the water diversion capacity of the lake is limited or the water quality cannot be changed with water diversion. Lake self-circulation can not only improve the water quality of the lake, but also avoids damage to the lake’s flow field.

Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/ijerph18115757/s1, Table S1: Calculation Programs and Results (W), Table S2: Calculation Programs and Results (K), Table S3: Calculation Programs and Results (Q), Table S4: Calculation Programs and Results (D). Int. J. Environ. Res. Public Health 2021, 18, 5757 15 of 16

Author Contributions: M.P. drafted the manuscript. W.S. carried on the model calculation. Y.L. made the contributions on the data analyses. Y.P. carried out the design of the outline. All authors have read and agreed to the published version of the manuscript. Funding: The research was supported by National Natural Science Foundation of China (Grants No. 5187090164), Priority Academic Program Development of Jiangsu Higher Education Institutions, the World-Class Universities (Disciplines) and Characteristic Development Guidance Funds for the Cen- tral Universities. Priority Academic Program Development of Jiangsu Higher Education Institutions. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References 1. Zhu, Y.; Liang, Z.; Li, W. Watershed water environment pollution models and their applications: A review. J. Appl. Ecol. 2013, 24, 3012–3018. 2. Zhang, Y.; Zhang, T.; Chen, J. Research progress on present situation and countermeasure of antimony pollution in water and soil environment. Ecol. Environ. Sci. 2011, 20, 1373–1378. 3. Song, W.; Xu, Q.; Fu, X.; Wang, C.; Pang, Y.; Song, D. EFDC simulation of fishway in the Diversion Dahaerteng River to Danghe Reservoir, China. Ecol. Indic. 2019, 102, 704–715. [CrossRef] 4. Song, W.; Xu, Q.; Fu, X.; Zhang, P.; Pang, Y.; Song, D. Research on the relationship between water diversion and water quality of xuanwu lake, China. Int. J. Environ. Res. Public Health 2018, 15, 1262. [CrossRef][PubMed] 5. Wang, X. Famous water transfer project in river basin. Water Resour. Electr. Power 2004, 30, 1–25. 6. Li, Y.; Tang, C.; Wang, C.; Anim, D.; Yu, Z.; Acharya, K. Improved Yangtze River Diversions: Are they helping to solve algal bloom problems in Lake Taihu, China? Ecol. Eng. 2013, 51, 104–116. [CrossRef] 7. Song, W.; Pang, Y.; Li, Y.; Zhang, P.; Xu, Q.; Fu, X. Calculation of water environmental capacity and pollutant sharing rate with water diversion: A case study of Qinhuai River. Water Supply 2019, 19, 1026–1035. [CrossRef] 8. Song, W.; Fu, X.; Pang, Y.; Song, D.; Xu, Q.; Zhang, P. Research on water environment regulation of artificial playground lake interconnected Yangtze River. Int. J. Environ. Res. Public Health 2018, 15, 2110. [CrossRef][PubMed] 9. Wang, H.; Zhang, Z.; Liang, D.; Pang, Y.; Hu, K.; Wang, J. Separation of wind’s influence on harmful cyanobacterial blooms. Water Res. 2016, 98, 280. [CrossRef] 10. Wang, H.; Zhao, Y.; Liang, D.; Deng, Y.; Pang, Y. 30+ year evolution of Cu in the surface sediment of Lake Poyang, China. Chemosphere 2017, 168, 1604–1612. [CrossRef] 11. Pang, M.; Song, W.; Zhang, P.; Shao, Y.; Li, L.; Pang, Y.; Wang, J.; Xu, Q. Research into the Eutrophication of an Artificial Playground Lake near the Yangtze River. Sustainability 2018, 10, 867. [CrossRef] 12. Tong, C.F.; Lv, L.R.; Shao, Y.Y.; Hao, J.L. Calculation and analysis of water quality improvement caused by water diversion in Nanjing Qinhuai River. In Applied Mechanics and Materials; Trans Tech Publications Ltd.: Baech, Switzerland, 2013; Volume 256, pp. 2528–2532. 13. Liu, C.; Zheng, H. South-to-north Water Transfer Schemes for China. Int. J. Water Resour. Dev. 2002, 18, 453–471. [CrossRef] 14. Huang, J.; Pang, Y.; Zhang, X.; Tong, Y. Water Environmental Capacity Calculation and Allocation of the Taihu Lake Basin in Jiangsu Province Based on Control Unit. Int. J. Environ. Res. Public Health 2019, 16, 3774. [CrossRef] 15. Li, R.; Gao, L.; Pan, C.; Pang, Y. Detecting the mechanisms of longitudinal salt transport during spring tides in Qiantang Estuary. J. Integr. Environ. Sci. 2019, 16, 123–140. [CrossRef] 16. Wang, H.; Ji, F.; Pang, Y. Fluctuation of river network water environmental carrying capacity in a complicated river-lake system. Environ. Eng. Manag. J. 2018, 17, 1511–1519. [CrossRef] 17. Song, W.; Pang, Y.; Shi, X.; Xu, Q. Study on the rectification of forebay in pumping station. Math. Probl. Eng. 2018, 2018, 2876980. [CrossRef] 18. Niu, Y.; Yu, H.; Niu, Y.; Jiang, X.; Guo, X.; Pang, Y.; Xu, X. Isotopic fractionation of particulate organic matter and its biogeochemical implication in the littoral zone of Lake Taihu, China. Water Sci. Technol. 2017, 76, 2690–2697. [CrossRef][PubMed] 19. Wang, J.; Zhao, Q.; Pang, Y.; Li, Y.; Yu, Z.; Wang, Y. Dynamic simulation of sediment resuspension and its effect on water quality in Lake Taihu, China. Water Sci. Technol. Water Supply 2017, 17, 1335–1346. [CrossRef] 20. Wang, J.; Zhao, Q.; Pang, Y.; Hu, K. Research on nutrient pollution load in Lake Taihu, China. Environ. Sci. Pollut. Res. 2017, 24, 17829–17838. [CrossRef] 21. Niu, Y.; Jiao, W.; Yu, H.; Niu, Y.; Pang, Y.; Xu, X.; Guo, X. Spatial evaluation of heavy metals concentrations in the surface sediment of Taihu Lake. Int. J. Environ. Res. Public Health 2015, 12, 15028–15039. [CrossRef] 22. Niu, Y.; Niu, Y.; Pang, Y.; Yu, H. Assessment of heavy metal pollution in sediments of inflow rivers to Lake Taihu, China. Bull. Environ. Contam. Toxicol. 2015, 95, 618–623. [CrossRef] Int. J. Environ. Res. Public Health 2021, 18, 5757 16 of 16

23. Wang, J.; Pang, Y.; Li, Y.; Huang, Y.; Luo, J. Experimental study of wind-induced sediment suspension and nutrient release in Meiliang Bay of Lake Taihu, China. Environ. Sci. Pollut. Res. 2015, 22, 10471–10479. [CrossRef][PubMed] 24. Xie, R.; Pang, Y.; Luo, B.; Li, J.; Wu, C.; Zheng, Y.; Sun, Q.; Zhang, P.; Wang, F. Spatiotemporal variability in salinity and hydraulic relationship with salt intrusion in the tidal reaches of the Minjiang River, Fujian Province, China. Environ. Sci. Pollut. Res. 2017, 24, 11847–11855. [CrossRef] 25. Zhang, P.; Pang, Y.; Shi, C.; Wang, Y.; Xu, L.; Pan, H.; Xie, R. Analysis and numerical simulation of natural and human-caused low dissolved oxygen in the Minjiang River Estuary. Water Sci. Technol. 2016, 73, 2475–2485. [CrossRef][PubMed] 26. Zhang, P.; Pang, Y.; Pan, H.; Shi, C.; Huang, Y.; Wang, J. Factors contributing to hypoxia in the Minjiang river estuary, Southeast China. Int. J. Environ. Res. Public Health 2015, 12, 9357–9374. [CrossRef][PubMed] 27. Wang, H.; Zhao, Y.; Zhang, Z.; Pang, Y.; Liang, D. Water transparency distribution under varied currents in the largest river- connected lake of China. Water Sci. Technol. 2017, 75, 406–417. [CrossRef][PubMed] 28. Omlin, M.; Brun, R.; Reichert, P. Biogeochemical model of Lake Zürich: Sensitivity, identifiability and uncertainty analysis. Ecol. Mod. 2001, 141, 105–123. [CrossRef] 29. Piermay, J.L. Long-term simulation of the eutrophication of the North Sea: Temporal development of nutrients, chlorophyll and primary production in comparison to observations. J. Sea Res. 1997, 38, 275–310. [CrossRef] 30. Nobre, A.M.; Ferreira, J.G.; Newton, A.; Simas, T.; Lcely, J.; Neves, R. Management of coastal eutrophication: Integration of field data, ecosystem-scale simulations and screening models. J. Mar. Syst. 2005, 56, 375–390. [CrossRef] 31. Scavia, D.; Powers, W.F.; Canale, R.P.; Moody, J.L. Comparison of first-order error analysis and Monte Carlo Simulation in time-dependent lake eutrophication models. Water Resour. Res. 1981, 17, 1051–1059. [CrossRef] 32. Geng, J.; Jin, X.; Wang, Q.; Niu, X.; Wang, X.; Edwards, M.; Glindemann, D. Matrix bound phosphine formation and depletion in eutrophic lake sediment fermentation-simulation of different environmental factors. Anaerobe 2005, 11, 273. [CrossRef] 33. Sagehashi, M.; Sakoda, A.; Suzuki, M. A mathematical model of a shallow and Eutrophic Lake (The Keszthely Basin, Lake Balaton) and simulation of restorative manipulations. Water Res. 2001, 35, 1675. [CrossRef] 34. Madden, C.J.; Kemp, W.M. Ecosystem model of an estuarine submersed plant community: Calibration and simulation of eutrophication responses. Estuaries 1996, 19, 457–474. [CrossRef] 35. Karim, M.R.; Sekine, M.; Ukita, M. Simulation of eutrophication and associated occurrence of hypoxic and anoxic condition in a coastal bay in Japan. Mar. Pollut. Bull. 2002, 45, 280–285. [CrossRef] 36. Huang, Y.L.; Huang, G.H.; Liu, D.F.; Zhu, H.; Sun, W. Simulation-based inexact chance-constrained nonlinear programming for eutrophication management in the Xiangxi Bay of Three Gorges Reservoir. J. Env. Manag. 2012, 108, 54. [CrossRef][PubMed] 37. Wang, J.; Pang, Y.; Li, Y.; Huang, Y.; Jia, J.; Zhang, P.; Kou, X. The regularity of wind-induced sediment resuspension in Meiliang Bay of Lake Taihu. Water Sci. Technol. 2014, 70, 167–174. [CrossRef][PubMed] 38. Marshall, H.G.; Burchardt, L. Phytoplankton composition within the tidal freshwater region of the James River, Virginia. Proc. Biol. Soc. Wash. 1998, 111, 720–730. [CrossRef] 39. Lung, W.S.; Paerl, H.W. Modeling blue-green algal blooms in the lower Neuse River. Water Res. 1988, 22, 895–905. [CrossRef] 40. Steinberg, C.; Hartmann, H. Planktonic bloom forming cyanobacteria and the eutrophication of lakes and rivers. Freshw. Biol. 1988, 20, 279–287. [CrossRef] 41. Eldridge, M.; Sieracki, M.E. Biological and hydrodynamic regulation of the microbial food web in a periodically mixed estuary. Limnol. Oceanogr. 1993, 38, 1666–1679. [CrossRef] 42. Qin, B. Approaches to Mechanisms and Control of Eutrophication of Shallow Lakes in the Middle and Lower Reaches of the Yangze River. J. Lake Sci. 2002, 14, 193–202. [CrossRef] 43. Qin, B.; Wang, X.; Tang, X.; Feng, S.; Zhang, Y. Drinking Water Crisis Caused by Eutrophication and Cyanobacterial Bloom in Lake Taihu: Cause and Measurement. Adv. Earth Sci. 2007, 22, 896–906. [CrossRef] 44. Song, W.; Pang, Y. Research on the influence of diversion flow ratio on the water environment in Nanjing. Water Supply 2019, 19, 2367–2378. [CrossRef] 45. Zhao, Z.; Mi, T.; Xia, L.; Yan, W.; Jiang, Y.; Gao, Y. Understanding the patterns and mechanisms of urban water ecosystem degradation: Phytoplankton community structure and water quality in the Qinhuai River, Nanjing, China. Environ. Sci. Pollut. Res. 2013, 20, 5003–5012. [CrossRef][PubMed] 46. Francisco, G.; Márcia, C.; Maria, F. Seasonal changes and spatial variation in the water quality of a eutrophic tropical reservoir determined by the inflowing river. Lake Reserv. Manag. 2011, 27, 343–354. [CrossRef] 47. Joyce, Z.; Patricia, B.; Nenad, I.; Steven, H.; Susan, G.; Richard, B. Long-Term Water Quality Trends in the Lake Okeechobee Watershed, Florida. Crit. Rev. Environ. Sci. Technol. 2011, 41, 548–575. [CrossRef] 48. Neyva, R.; Rigoberto, C.; Eliandro, C.; Paulo, N. Local calibration of coliforms parameters of water quality problem at Igapó I Lake, Londrina, Paraná, Brazil. Ecol. Model. 2011, 222, 1888–1896. [CrossRef] 49. Raymond, C.; Dong-Il, S. Performance, reliability and uncertainty of total phosphorus models for lakes—II. Stochastic analyses. Water Res. 1996, 30, 95–102. [CrossRef]