The connectivity of urban river networks plays an important role in cities in many aspects, such as urban water safety, water quality (WQ), and aquatic ecological balance. This study focuses on the river network and the Majiawan Wetland in the Chaoyang District of Beijing by establishing a two-dimensional hydrological WQ model employing various water allocation schemes between the river network and the wetland. Water circulation and WQ are the main indexes, and the effects of different scenarios on improving water circulation and WQ are simulated and compared. This study demonstrates that the addition of water replenishment at the intersection of river network and internal slow-water zones of the wetland (Scheme 2) has greater effectiveness in improving both hydrology and WQ compared to two other schemes. The water area of the Majiawan Wetland has expanded, and water velocity has increased. Using chemical oxygen demand, total nitrogen, and total phosphorus as the index values for determining the water class, the WQ of about 20% of the wetland area was reached Water Class II (domestic drinking water), with Water Class III (general industrial water) accounting for the other 80%. This study provides valuable evaluation and reference for similar areas of urban river network connectivity.

  • This study established a comprehensive hydrological and water quality model for the urban river network-wetlands system.

  • We simulated and analyzed the water cycle allocation schemes of the urban river network-wetlands from the water allocation, water quality index changes and environmental impact evaluation, by using the established mathematical model.

  • The research results would offer valuable insights for optimizing water resource allocation and improving water quality in the study area and similar water systems.

In recent decades, the development of cities and urban settlements has escalated continuously with the rapid advancement of urbanization. The construction of urban waterways and wetland parks has been increasing, leading to a complex and diverse urban river network system (Wang et al. 2018). However, it is inevitable that waste and wastewater from residential and industrial activities may flow into the urban river network, conceivably causing environmental changes in water quality (WQ) and sedimentation issues in the river channels. Such wastewater could also result in the blockage of urban river networks, affecting the WQ of river channels and wetlands. This, in turn, can cause damage to the overall urban WQ. Thereby, the sustainable development of cities and urban settlements might be negatively impacted (Liu 2017; Wang 2018).

In China, water system connectivity has become an important water management strategy. It aims to achieve a more integrated and streamlined river network system by overcoming issues, such as poor water flow in traditional water management methods and the degradation of the ecological environment. Due to the severe water resource situation, water system connectivity has gained increasing attention as a key approach in modern water network construction and an essential component of Sponge City development. It is considered an advanced method for the ecological restoration of water systems (Meng et al. 2014; Zhou et al. 2021). Various methods have been developed to evaluate the connectivity of urban river networks, including graph theory; hydrological and hydraulic methods, landscape approaches, biological methods; and comprehensive index methods (Yan & Li 2010; Cui et al. 2011). These methods integrate knowledge from different disciplines' and technical approaches, providing diverse tactics for assessing water system connectivity.

Numerous scholars have been continuously delving into the mechanistic understanding of the impact of human interventions on hydrodynamics and WQ conditions. Additionally, numerical simulations play a crucial role in analyzing the constructive and adverse effects of hydraulic structures on water bodies (Zhang et al. 2011; Quinlan et al. 2015). The repercussions of human interventions are highly intricate. However, it is evident that the assimilative capacity of riverine water bodies can be enhanced through effective hydraulic regulation, and WQ can be significantly improved (Hachikyan et al. 2005). Rational hydraulic intervention measures help to decrease the concentrations of pollutants (such as chemical oxygen demand (COD), total phosphorus (TP), and total nitrogen (TN)), while a dilution effect also aids in achieving this objective (Lu et al. 2019). This comprehensive and effective hydraulic control holds profound significance in maintaining ecological balance in river channels and promoting the sustainable utilization of water resources (Liu et al. 2021).

This study employs the hydraulic WQ method to investigate the river network–wetland system in the Majiawan Wetland, Chaoyang District of Beijing. A comprehensive mathematical model is developed to analyze the hydrology and WQ of the urban river network–wetland. Multiple scenarios for water system connectivity scheduling are designed, and the variations in hydraulic and WQ parameters were simulated and calculated under different operational conditions. The primary objective of this study is to explore optimized scheduling strategies for enhancing the connectivity between urban river networks and wetlands. The research outcomes of this study provide scientific support for the efficient allocation of water resources in the research area and similar water systems. The results also have significant practical value in applications such as hydrodynamic enhancement; the improvement of water circulation in river network–wetlands; and the prevention and control of water pollution.

Study area

The study area is located in the southern river network–wetland system of the Chaoyang District of Beijing. The total area of the study area is around 90 km2, spanning 9 km from east to west and 11 km from north to south. The eastern boundary starts from the upstream of the Xiaotaihou River, while the western boundary extends to the Majiawan Wetland. The northern boundary begins at the source of the Tonghui River Irrigation Channel, and the southern boundary reaches the Majiawan Wetland. The study area includes various waterways such as the Halfbidian River, the Daliushu River, the Guanyintang River, the Xianninghou River, the Dagao River, the Erzhi Canal, and the Nanda River. These waterways constitute the urban river network–wetland system in the study area, as shown in Figure 1.
Figure 1

Location map of the Southern Water System–Wetland in the Chaoyang District.

Figure 1

Location map of the Southern Water System–Wetland in the Chaoyang District.

Close modal

Numerical simulation and calculation

Introduction to the mathematical model

In order to better fit the river channel boundaries and accurately simulate the impact of the water system connectivity scheduling measures on the hydraulic and WQ conditions of the Majiawan Wetland in the Chaoyang District, this study utilizes the MIKE 21 model to establish a two-dimensional hydrological and WQ model for the river network–wetland system. This model is based on the finite volume method and uses a numerical two-dimensional unstructured grid model. It can employ triangular grids, rectangular grids, or a combination of triangular and rectangular grids to discretize the simulation area. Additionally, it allows for the local refinement of boundaries and key areas of interest, exhibiting characteristics such as discrete simplicity and good conservation.

Model control equations

The basic equations of incompressible fluid motion are used to derive shallow water equations by neglecting the variations of physical quantities along the water depth and integrating along the water depth direction (Xu 2010; Chang et al. 2021).

Continuity equation:
formula
(1)
Equations of motion:
formula
(2)
formula
(3)
where η is the water level relative to the reference datum; h is the water depth; t is the time; ρ is the density of water; g is the acceleration due to gravity; C is the Chezy coefficient; f is the Coriolis force coefficient; represent the wind-induced stress; and Ex and Ey are the generalized eddy viscosity coefficients in the x and y directions, respectively.
Building upon the hydrodynamic model, considering pollutant decay and diffusion, a pollutant WQ migration and diffusion model is developed:
formula
(4)
where C is the concentration of pollutants in the water environment; t is time; u and v are the velocity components in the x and y directions, respectively; D is the horizontal diffusion coefficient; and k represents the primary decay coefficient of the substance.

In this study, the values of model parameter k for the pollutant indicators are determined based on the existing data and relevant WQ research results in the river network–wetland as follows: the COD decay coefficient is 0.01/d; the TN decay coefficient is 0.08/d; and the TP decay coefficient is 0.004/d (Liang et al. 2003; Wu & Dong 2008).

Model domain and grid

In this study, the Majiawan Wetland serves as the primary focus, and the hydrodynamic conditions of selected neighboring waterways are considered by integrating their connectivity into the computational domain. These waterways include the Tonghuiguanqu River, the Tonghuipaigan River, the Daliushu River, the Xiaotaihou River, and the Majiawan Wetland, as depicted in Figure 2. To accurately simulate the hydrodynamics of both meandering and straight waterways, a quadrilateral grid discretization method is employed. For extensive wetland areas, a triangular grid discretization method is utilized. Additionally, local refinement is applied at the confluence points of the Majiawan Wetland and various waterways to enhance the model's stability and precision. The model spans approximately 9 km north to south and 11 km east to west, comprising a total of 20,851 grids and 19,106 grid nodes. The grid sizes range from approximately 1 to 15 m, as illustrated in Figures 2 and 3.
Figure 2

The scope location map of the model.

Figure 2

The scope location map of the model.

Close modal
Figure 3

Grid discretization of the local region (Majiawan Wetland) in the computational model.

Figure 3

Grid discretization of the local region (Majiawan Wetland) in the computational model.

Close modal

Model parameters and conditions

Manning's roughness coefficient (n)

Determining model parameters based solely on existing data is challenging. Therefore, the roughness coefficient refers primarily to the natural roughness of the river. In this study, the roughness coefficient for the channels is set at 0.02, based on prior research results related to wetlands in the river network (Liang et al. 2003; Wu & Dong 2008).

Horizontal eddy viscosity coefficient
The horizontal eddy viscosity coefficient is calculated using the Smagorinsky model:
formula
(5)
formula
(6)
where Cs is a constant, and in this study, it is set to 0.28; L represents the characteristic length; and Sij represents the deformation rate.
Wet and dry boundary conditions

The wet–dry point discrimination method is adopted to handle the dynamic boundary water area. In order to improve the stability of the model calculation, this study sets the wet and dry boundary conditions, as seen in Table 1.

Table 1

Wet and dry boundary conditions

TypeWet and dry boundary conditions (m)
Dry depth condition 0.005 
Wet depth condition 0.05 
Flooding depth condition 0.001 
TypeWet and dry boundary conditions (m)
Dry depth condition 0.005 
Wet depth condition 0.05 
Flooding depth condition 0.001 

Initial conditions

Hydrodynamic initial conditions: The stabilized flow field of the model trial calculation is used as the initial conditions.

WQ initial conditions: In this study, the initial concentration of pollutants was determined based on Water Class V indicator values, which are as follows: COD is 40 mg/L, TN is 2.0 mg/L, and TP is 0.4 mg/L, as seen in Table 2.

Table 2

Water quality indicators of water classes

TypeWater quality indicators (mg/L)
CODTNTP
Water Class II 15–20 0.5–1.0 0.1–0.2 
Water Class III 20–30 1.0–1.5 0.2–0.3 
Water Class IV 30–40 1.5–2.0 0.3–0.4 
Water Class V ≥40 ≥2.0 ≥0.4 
TypeWater quality indicators (mg/L)
CODTNTP
Water Class II 15–20 0.5–1.0 0.1–0.2 
Water Class III 20–30 1.0–1.5 0.2–0.3 
Water Class IV 30–40 1.5–2.0 0.3–0.4 
Water Class V ≥40 ≥2.0 ≥0.4 

Boundaries on both sides of the river

Land boundary conditions are used.

Water level and flow boundary of the river

The flow control is adopted in the upstream of the river, and the water level control is adopted in the downstream, which can be found in the river network–wetland hydraulic conditions, as seen in Table 3 of design schemes.

Table 3

Design specification of the schemes

Scheme descriptionSpecific design
Scheme 1 Based on the current situation of the water system–wetland, the aim is to increase water intake and replenishment of the wetland WaterIntakePoint 2 (64,800 m3/d) for water intake and WaterSupplyPoint 3 (64,800 m3/d) for water replenishment, with an operating water level of 26 m 
Scheme 2 Based on Scheme 1, water intake is increased at the intersection of the water system river network, and internal water replenishment systems are established within the wetland Based on Scheme 1, add WaterIntakePoint 1 (100,000 m3/d) for water replenishment, WaterSupplyPoint 2 (100,000 m3/d) for water intake, and WaterSupplyPoint 4 (20,000 m3/d) for water replenishment, with an operating water level of 26 m 
Scheme 3 Based on Scheme 2, the water level of the Majiawan Wetland is lowered Based on Proposal 2, the operating water level is lowered to 25.7 m 
Scheme descriptionSpecific design
Scheme 1 Based on the current situation of the water system–wetland, the aim is to increase water intake and replenishment of the wetland WaterIntakePoint 2 (64,800 m3/d) for water intake and WaterSupplyPoint 3 (64,800 m3/d) for water replenishment, with an operating water level of 26 m 
Scheme 2 Based on Scheme 1, water intake is increased at the intersection of the water system river network, and internal water replenishment systems are established within the wetland Based on Scheme 1, add WaterIntakePoint 1 (100,000 m3/d) for water replenishment, WaterSupplyPoint 2 (100,000 m3/d) for water intake, and WaterSupplyPoint 4 (20,000 m3/d) for water replenishment, with an operating water level of 26 m 
Scheme 3 Based on Scheme 2, the water level of the Majiawan Wetland is lowered Based on Proposal 2, the operating water level is lowered to 25.7 m 

River network–wetland hydraulic conditions
This study focuses on the water system in the southern part of the Chaoyang District and the Majiawan Wetland. Based on the existing river topography and section flow data, the design includes internal flow regulation information for the river network and the wetland. The major components of the river system and the wetland are as follows: incoming water volume in the upstream of the Xiaotaihou River (60,000 m3/d); water supply point one in the upstream of the Xiaotaihou River (64;800 m3/d); the Tonghuipaigan River (20;000 m3/d); the Tonghuiguanqu River (51;000 m3/d); the Daliushugou River (19;000 m3/d); the Banbidian River (0.16 million m3/d); the Xianninggou River (8;000 m3/d); the Dagaogou River (76;000 m3/d); the Erzhiqu River (62;000 m3/d); the Nandagou River (8;000 m3/d); and the Majiawan Wetland (with a circulating replenishment pump station of 64,800 m3/d). The location information of related river network flow replenishment points can be seen in Figure 4.
Figure 4

Distribution map of river network flow replenishment points.

Figure 4

Distribution map of river network flow replenishment points.

Close modal

Different schemes

The aim of this study is to investigate the hydraulic adjustment and WQ changes in the Majiawan Wetland. The background WQ is set according to Water Class V standards, and all water supply points meet Water Class II standards as required. Various water system scheduling schemes are designed, and the WQ indicators including COD, TN, and TP are evaluated based on the standards set in the ‘Surface Water Environmental Quality Standards (GB3838-2002)’. Specific details of the Water Classes can be seen in Table 2 (Zhu et al. 2009). The study simulates and calculates the WQ changes after the water system adjustment from Class V WQ through the interconnected water system.

Based on the hydraulic information, WQ, and river channel distribution of the river–wetland system, three interconnected water system scheduling schemes are designed. Simulation predictions are conducted for different scheduling schemes, focusing on the water flow dynamics and WQ changes in the Majiawan Wetland. The specific details of the schemes can be seen in Table 3 and Figure 4.

Based on the hydraulic and WQ status of the river–wetland system, this study has analyzed the hydrological and WQ changes of three different scheduling schemes. The results show that areas with high water velocity in the Majiawan Wetland are found mostly in narrow river channels and confluence areas of the wetland entrances and exits. Most of the channel widths of the Majiawan Wetland network range between 30 and 50 m. In the southern area of the Majiawan Wetland, the channel widths are between 10 and 25 m. According to the simulation calculation analysis of the three schemes, the average water velocity of the Majiawan Wetland network is 0.03 m/s, the minimum water velocity is 0.1 m/s, and the maximum water velocity is 0.15 m/s, as shown in Figure 5. The internal circulation between the wetland and connecting waterways can increase flow intensity, resulting in higher water velocity and accelerated water circulation throughout the Majiawan Wetland network. This, in turn, changes the water environment quality in the entire Majiawan Wetland network, with the potential for a maximum improvement from Water Class V to Water Class II. With no change in the internal circulation measures, the water level of the Majiawan Wetland is lowered by setting an overflow weir downstream of the river network–wetland model. This reduces the watershed area and the reservoir capacity of the wetland but speeds up the overflow of the Majiawan Wetland network. It also enhances the water movement and intensity of water circulation for the overall Majiawan Wetland network, promoting the quality of the water environment. In addition, in the stream area in the southern area of the Majiawan Wetland, the rapid circulation of water in the open wetland area results in the rapid pumping of water into the stream area. At the same time, due to the high topography of the southern stream, the phenomenon of water flow truncation occurs, which reduces water flow velocity, diminishes the intensity of water circulation, and degrades the capability for water purification.
Figure 5

Distributions of water surface flow lines and water velocities in the Majiawan Wetland under different schemes.

Figure 5

Distributions of water surface flow lines and water velocities in the Majiawan Wetland under different schemes.

Close modal

Hydrodynamic analysis

Scheme 1 considers the existing condition of the water system in the Majiawan Wetland, and it incorporates water supply measures at the ingress and egress points of the wetland to improve water availability. The simulation results of Scheme 1 showed that the water velocity increases at the main stream lines of the wetland, with a maximum velocity of 0.1 m/s in narrow areas of the river network. However, the stream area in the southern area of the Majiawan Wetland has very little water storage, the intensity of water movement is very low, and most of the water flow velocity is less than 0.01 m/s, as shown in Figure 5.

Building on Scheme 1, Scheme 2 adds water supply measures at the southern streams of the wetland. Moreover, it adds water intake measures at the intersection of the Majiawan and Tonghui Drain. As compared to Scheme 1, the simulation results of Scheme 2 showed that the water velocity further increases at the entrances and exits of the wetland, with a maximum velocity of 0.15 m/s in narrow areas of the river network. The water quantity and flow intensity in the southern part of the wetland increase better, with most velocities increasing by more than 0.02 m/s. The overall flow activity and water exchange capacity of the wetland increase better, as shown in Figure 5.

Scheme 2 and Scheme 3 had the same water system scheduling measures, differing only in the water level in the wetland. The simulation results of Scheme 3 showed that the water velocity increases at the entrances and exits of the wetland, with a maximum velocity greater than 0.13 m/s in narrow areas of the river network. Due to the lower water level, the water area in the southern part of the wetland decreases better, leading to an increase in slow-flowing areas, as shown in Figure 5.

Overall, Scheme 2 improved the water flow dynamics of the Majiawan Wetland, increased the water surface area, and reduced the slow-flow areas within the wetland. The simulation results of Scheme 2 were better than those of the other schemes.

WQ analysis

The simulation calculated the variation patterns of WQ indicators in the Majiawan Wetland under different flow conditions of different schemes. The specifics of the three schemes were as follows.

Scheme 1

For the simulation results of Scheme 1, the WQ of the whole Majiawan Wetland could reach from Water Class V to Water Class II or Water Class III. At the same time, the average removal efficiency of COD could reach 37%; the average removal efficiency of TN could reach 45%; and the average removal efficiency of TP could reach 30%.

For the pollution indicator value COD to determine the Water Class, around 80% of the wetland area could become Water Class III and around 20% of the wetland area could attain Water Class IV, mainly at the southern area of the wetland. For the pollution indicator values TN and TP to determine the Water Class, around 50% of the wetland area could become Water Class II, with 28% being Class III and 22% being Class IV, as shown in Figure 6.
Figure 6

Distributions of COD, TN, and TP concentrations in the Majiawan Wetland in Scheme 1.

Figure 6

Distributions of COD, TN, and TP concentrations in the Majiawan Wetland in Scheme 1.

Close modal

Scheme 2

For the simulation results of Scheme 2, when compared with Scheme 1, the water velocity and water circulation intensity of the whole Majiawan Wetland were further enhanced. As a result, the WQ of the entire Majiawan Wetland could reach Water Class III or Water Class II, whereas previously it was classified as Water Class V. The average removal efficiency of the pollution index value COD could reach 45%; the average removal efficiency of the pollution index value TN could reach 55%; and the average removal efficiency of the pollution index value TP could reach 39%.

For the pollution indicator value COD to determine the Water Class, around 60% of the wetland area was classified as Water Class III, around 15% was classified as Water Class IV, and around 25% was classified as Water Class II, which was mainly concentrated in the southern streams of the wetland. For the pollution indicator values TN and TP, around 35% of the wetland became Water Class II, with around 65% being Water Class III. Overall, the water velocity and water circulation intensity in the wetland increased, leading to improved WQ, as shown in Figure 7.
Figure 7

Distributions of COD, TN, and TP concentrations in the Majiawan Wetland in Scheme 2.

Figure 7

Distributions of COD, TN, and TP concentrations in the Majiawan Wetland in Scheme 2.

Close modal

Scheme 3

For the simulation results of Scheme 3, when compared with Scheme 2, the water level of the whole Majiawan Wetland decreased, the water area and reservoir capacity decreased, and the water velocity was further enhanced. However, water truncation occurred in the stream waters of the high topography in the southern part of the wetland. The WQ of the entire Majiawan Wetland could reach Water Class II or Water Class III from Water Class V. The average removal efficiency of the pollution indicator value COD could reach 48%; the average removal efficiency of the pollution indicator value TN could reach 58%; and the average removal efficiency of the pollution indicator value TP could reach 42%.

For the pollution indicator value COD to determine the water class, around 10% of the wetland area was Water Class IV in the stream waters of the southern wetland area; around 75% was Water Class III; and around 15% was Class II. For the pollution indicator values TN and TP, around 90% of the wetland became Water Class II and III, with around 10% being Water Class IV in the stream waters of the southern wetland area, as shown in Figure 8.
Figure 8

Distributions of COD, TN, and TP concentrations in the Majiawan Wetland in Scheme 3.

Figure 8

Distributions of COD, TN, and TP concentrations in the Majiawan Wetland in Scheme 3.

Close modal

All in all, Scheme 2 was more effective than the other two schemes in improving the WQ. The WQ of the wetland could reach Water Class II and III, with Water Class III accounting for about 80% of the wetland water area and Water Class II accounting for the other 20%. Meanwhile, the WQ in the stream waters of the southern wetland area was also effectively improved. This comprehensive improvement has had a positive impact on the overall WQ status of the Majiawan Wetland.

This study aimed to investigate the hydraulic adjustment and WQ changes in the Majiawan Wetland. Three interconnected water resource scheduling schemes were designed based on the hydraulic condition, WQ, and river channel distribution of the river–wetland system. Based on those three water resource scheduling schemes, the water flow cycle and WQ distribution of the river network–wetlands were simulated and analyzed. The research findings provide valuable insights for optimizing water allocation and improving WQ in the study area and similar areas. The main conclusions are as follows:

  • (1)

    Analysis of the three water resource scheduling schemes revealed variations in water surface velocity and WQ distribution. Water resource scheduling can enhance wetland water flow dynamics and improve WQ. Increasing the implementation of water supply measures effectively enhances the overall water flow intensity and improves the WQ status. The minimum water velocity is less than 0.01 m/s, which is observed mainly in areas with smaller water volume. While the maximum water velocity exceeds 0.1 m/s, primarily occurring in narrow sections of the river–wetland system or along the main stream lines.

  • (2)

    Scheme 2 proved to be more effective in improving hydrology and WQ compared to the other two schemes. The simulation results demonstrate an overall enhancement in the flow dynamics of the Majiawan Wetland, with increased water velocity and improved water exchange capacity. In this study, COD, TN, and TP are used as the index values for determining the water class levels, and the WQ improvement of the Majiawan Wetland was shown to reach Water Class III or close to Water Class II. Approximately 80% of the wetland area can achieve Water Class III, while around 20% can achieve Water Class II. Notably, significant improvement in WQ was also observed in areas with slower water flow. Overall, hydrological dynamics and WQ status can be significantly enhanced.

There is no specific funding to support this research.

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

The authors declare there is no conflict.

Chang
Z.
,
Zhang
Q. H.
,
Jiang
Y.
,
Sun
N.
,
He
R.
,
Wang
T. F.
,
Wu
Y. J.
&
Long
X. F.
2021
Analysis of the impact of sewage outfall on the water quality of coastal waters of Macau based on MIKE21 model
.
Pearl River
42, 3.
Cui
G. T.
,
Zuo
Q. T.
&
Duo
M.
2011
Development evolution and influences of the interconnected river system network at home and abroad
.
South-to-North Water Diversion and Water Science & Technology
9
,
4
.
Hachikyan
A.
,
Scherer
R.
,
Angelova
M.
&
Lazarova
M.
2005
Decision support system for a river network pollution estimation based on a structural-linguistic data model
.
Operational Research
5
,
105
113
.
Liang
J.
,
Zhou
Q.
&
Sun
T.
2003
A research review and technical improvement analysis of constructed wetland systems for wastewater treatment
.
Chinese Journal of Ecology
22
,
49
55
.
Liu
J.
2017
Functional evaluation of water system connectivity in Chaoyang City
.
Water Resources Planning and Design
4
, 13.
Liu
Z. Q.
,
Wang
H.
,
Feng
X. Y.
,
Yan
H. Y.
&
Xia
K.
2021
Values of plain river network hydraulic regulatory threshold based on water quality objectives
.
Environmental Science & Technology
44, 5.
Lu
Y. W.
,
Pang
Y.
&
Zhou
R. R.
2019
Research on the water diversion scheme on water environment improvement in the Yundong District of Wuxi City in Taihu Basin
.
Sichuan Environment
38, 1.
Meng
X. Y.
,
Chen
X.
,
Chen
D. Y.
,
Hang
Q.
&
Zhu
Y.
2014
Evaluation system of urban water system connectivity
.
Journal of Hohai University (Natural Sciences)
42
,
5
.
Quinlan
E.
,
Gibbins
C. N.
,
Batalla
R. J.
&
Vericat
D.
2015
Impacts of small scale flow regulation on sediment dynamics in an ecologically important Upland River
.
Environmental Management
55
(
3
),
671
686
.
Wang
X.
2018
Study on Interconnected River Network System and Hydrodynamic Water Environment in Haikou City
.
Master's dissertation. South China University of Technology, Guangzhou, China
.
Wang
Q. G.
,
Wang
Y. P.
,
Lu
X. C.
,
Jia
P.
,
Zhang
B. B.
,
Li
C.
,
Li
S.
&
Li
S. B.
2018
Impact assessments of water allocation on water environment of river network: method and application
.
Physics and Chemistry of the Earth
103
,
101
106
.
Wu
S. B.
&
Dong
R. J.
2008
The application and research progress of constructed wetland for wastewater treatment
.
Technology of Water Treatment
8, 5–9.
Xu
T.
2010
Calculation principle and application example of a two-dimensional flow model-MIKE21 HD
.
Water Conservancy Science and Technology and Economy
47
,
5
.
Yan
J. Y.
&
Li
N. G.
2010
Water network connection project of Donghu Lake in Wuhan City
.
Yangtze River
41, 11.
Zhang
Q.
,
Chen
Y. D.
,
Jiang
T. X.
&
Liu
Z.
2011
Human-induced regulations of river channels and implications for hydrological alterations in the Pearl River Delta, China
.
Stochastic Environmental Research and Risk Assessment
25, 7.
Zhou
Y. J.
,
Shi
Y. N.
,
Wu
H.
,
Chen
Y. F.
,
Yang
Q.
&
Fang
Z. S.
2021
Evaluation and simulation optimization of river network connectivity using graph theory and network flow
.
Water Resources and Power
39, 1.
Zhu
X. W.
,
Wang
X. P.
,
Li
Q.
,
Sha
J.
&
Wang
Y. Q.
2009
Environmental Quality Standards for Surface Water (GB 3838-2002): Practice and Recommendations
. In
Proceedings of the 2009 Annual Conference of the Chinese Society for Environmental Science
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/).