Abstract

It is difficult to simultaneously manage the dynamic demands for river water quality and quantity, and reconcile the contradiction between socio-economic and eco-environmental water consumption. As a solution, we proposed a three-tier model to optimize the distribution of river water resources. Using three constraint conditions, namely the ratio of wastewater to clean water, the eco-environmental water requirements of each node and the use of wetland projects, we determined various water quantity and quality allocation scenarios. We tested the model on the Qingyi River, and found that, for the recommended scenario that involved enhanced water-saving, a wastewater/clean water ratio and wetlands, more than 80% of the eco-environmental water requirements of each node except for the Luma section were achieved for each month. While the water quality in some individual river sections did not meet the standards for a few months of the year, the water quality of the remaining sections could be improved from class V, the current state, to class IV, and ammonia nitrogen and chemical oxygen demand (COD) standards could be achieved 82% and 96% of the time, respectively. These results show that the proposed model is effective and fit for purpose.

INTRODUCTION

Sufficient water is needed to maintain the ecological environmental functions of rivers (Song et al. 2003). In recent years, over-exploitation of water resources has resulted in eco-environmental problems. For example, in Canada, the United States, and Europe, almost 75% of the 85 largest rivers are affected by dams, which reduce the flow and destroy ecosystems (Hatry et al. 2013). The 2015 Environment Bulletin of the Chinese National Ministry of Environmental Protection reported that water quality was worse than the Class V level of the Environmental Quality Standards for Surface Water across 13.6% of the state-controlled sections of 10 major basins.

The aim of optimal allocation of water resources is to achieve a balance between the economy, ecology, society, and green development of water resources systems (Wang et al. 2016). Since the 1970s, with aggravation of the water crisis and pollution, water quality has been an increasingly important constraint on the water resources allocation process. Pingry et al. (1990), in an attempt to achieve a balance between water resources allocation and water pollution in the Colorado River, built a decision support system that combined water quality and quantity. Carlos et al. (1997) developed the optimal multiple water resources allocation model that considered the water quality requirements of different water users that gave maximum economic benefit. Avogadro et al. (1997) and Zhang et al. (2010) coupled water quantity and water quality in their model, respectively. They used the model to determine whether the results from water quality simulations met the water quality target for pollutant reduction.

In China, the Ministry of Water Resources of the People's Republic of China proposed that the eco-environmental water demand should be considered in water resources planning and allocation in the late 1990s (Liu 2002). Through a series of studies over recent years, the methods for calculating the eco-environmental water demand (Penas et al. 2014; Meng et al. 2016) and regulating the ecological and environmental water requirements of rivers that are sustained by natural runoff (Gao et al. 2012) have improved. Studies of the integrated optimal distribution of water quantity and quality in polluted rivers with low flow are relatively novel. The ecological and environmental water requirements of a river need to be determined to ensure sufficient water of a suitable quality and river ecosystem health. Therefore, we used a three-tier coupled model for the optimal distribution of water resources to determine the optimal ecological and environmental water requirements of low-flow river with weak self-purification ability. The theory and methods for protecting both the water quantity and quality in low-flow polluted rivers provide the key to maintaining favorable eco-environmental water conditions.

METHODS

Construction of the three-tier model for the optimal distribution of water resources in low-flow polluted rivers

First-tier macro allocation model

  • (1)

    Objective function

In line with the total amount control principle, the optimal water resources allocation model was constructed to achieve maximum comprehensive benefit. Among them, the maximum GDP, the minimum total water shortage and the minimum eco-environmental water shortage of the basin represented the economic benefit target, the social benefit target and the eco-environment target, respectively (Hu 2012).

  • (2)

    Constraint conditions

    • Ratio of wastewater/clean water

      According to the balance between water quality and quantity, we have  
      formula
      (1)
      where and are the water quantity and quality of clean water, respectively, and and are the water quantity and the pollutant concentration of wastewater, respectively. C is the water quality of the mixed water of the clean and wastewater.
      Given that one of the objectives of the optimal distribution of water resources is to improve the river water quality, we have to ensure that the mixed water quality C is better than the water quality target .  
      formula
      (2)
      The ratio of wastewater to clean water is defined as  
      formula
      (3)
      Therefore, we have  
      formula
      (4)
    • Other constraints

      Other constraints include the water supply capacity and the water carrying capability of the basin, water demand, and the non-negative constraints of variables.

Second-tier model for optimizing the quantity of allocated river water

  • (1)

    Objective function

The eco-environmental water requirements of the river at different times and in different river reaches were used as the condition criteria. The multi-objective model for the optimal allocation of the river water was constructed to satisfy the eco-environmental water requirement as far as possible and minimize the loss of unconventional water sources.

  • Fairness objective: The aim was to achieve the maximum overall satisfaction of water quantity for each time period and control node (Peng et al. 2016).  
    formula
    (5)
    where E is the overall satisfaction with the water quantity, k is the computing unit, is the total number of computing units, m is the control node, is the total number of water allocation nodes, t is time, is the monthly period, t = 1, 2,… 12, and is the water allocated to the control node m in the time t in the computing unit k. The water demand is .
  • Economic objective: the aim was to minimize the loss of water from unconventional water sources for each time period and control node.  
    formula
    (6)
    where a is the allocated water from the unconventional water sources, n is the allocated water to the eco-environmental water demand of the river system, is the water allocated to water user n in the computing unit k from the water source a.
  • (2)

    Constraint conditions

    • Water demand of the river control nodes. The water allocated to each node should not be less than the basic ecological water demand, and should not exceed the maximum water requirement.  
      formula
      (7)
    • Water-carrying capability of the water conveyance system. The water allocated to each node should not exceed the maximum water-carrying capability of the river's water transmission pipeline, which is supplied from corresponding water source.  
      formula
      (8)
    • Operation scale of pollutant reduction projects. In order to make the river water quality meet the standard, the effluent from sewage treatment plants enters the pollutant reduction project for advanced treatment. The scale of pollutant reduction project should not exceed the design scale.  
      formula
      (9)
    • Non-negative constraints of variables.

      Where is the basic ecological water demand of the control node m, is the maximum water requirement for the control node m for time t, is the maximum water-carrying capability of the control node m for time, is the operation scale of the pollutant reduction projects, is the design scale of the pollutant reduction project.

  • (3)

    Model solution

The first- and second-tier models were solved by Linear Interactive and General Optimizer (LINGO), which was developed by LINDO Systems and is used to solve optimization problems. Because of its computational stability, rapid processing, and simple editing, it is suitable for solving linear and nonlinear optimization problems. The computer implementation process which solves the first-tier model and second-tier model optimal allocation model includes data input, software operation and result output. The input data is supported by the excel files, which include the main data such as water demand capacity forecast, water supply capacity forecast, water delivery capacity and intermediate data in order to realize the multiple target and nondimensionalization, the optimal allocation results in water quantity.

Third-tier water quality simulation model

The structure of the hydrodynamic water quality model was based on the Saint-Venant equation of one-dimensional flow and the general equation for one-dimensional water quality. The parameters were scalibrated and the water quality was simulated using 2007 Edition MIKE 11.  
formula
(10)
 
formula
(11)
where A is the wetted area, t is time, Q is flow, x is distance, q is the flanking inflow, α is the momentum correction coefficient, h is water level, g is acceleration due to gravity, C is Chezy's coefficient, R is the hydraulic radius, is the concentration, D is the diffusion coefficient, K is the degradation coefficient, and is the concentration of the source/sink term.

Framework of the allocation model

We established the three-tier model for the optimal distribution of river water resources. The first-tier model coordinates the contradictions between the social and economic water demands of the river basin, and the river ecological and environmental water requirements. The second-tier model determines the river ecological and environmental water quantity requirements for each time step and control node, and the third-tier model assesses the river water quality. The output of the previous tier is the input for the next tier.

We developed the set of water quantity and quality allocation scenarios. Among them, water saving in the socio-economic system is referred to in the water quantity allocation scenario. If the river ecological and environmental water requirements cannot be met by the allocated water, water saving measures will be added and the allocation of water resources in the whole basin will be re-optimized. The water quality allocation scenario can be changed by adjusting the allocation parameters (R) and adding pollution reduction measures. The parameters can be adjusted to ensure the water quality meets the river eco-environmental water requirements, and the three-tier model is optimized to ensure the water demands from society and the economy are met. If the water quality still fails to meet the requirements, the quantity of pollutants discharged to the river is reduced by pollution reduction measures, and then the optimal distribution of the water resources is calculated by the second and third tiers. The model framework is shown in Figure 1.

Figure 1

Optimal distribution of water resources in the rivers based on the eco-environmental water requirements.

Figure 1

Optimal distribution of water resources in the rivers based on the eco-environmental water requirements.

EXAMPLE ANALYSIS

Background to the study region

The Qingyi River is part of the Huaihe River Basin (Figure 2). It originates in Xinzheng City, enters Xuchang City from Guanting in Changge City, a county-level city, and flows into the Yinghe River at Taocheng Gate in Yangling County. It receives industrial and domestic wastewater along its entire course, and rainfall is the only natural water source. In the Qingyi River, chemical oxygen demand (COD) levels are much higher than the standard while ammonia nitrogen concentrations exceed the standard across the entire basin. The Qingyi River, characterized by a high degree of human disturbance and severe ecosystem degradation (Chang et al. 2015), is a good example of a low-flow polluted river.

Figure 2

Water system map of the Qingyi River basin.

Figure 2

Water system map of the Qingyi River basin.

Defining the key problems of water resources allocation

We selected 2020 as the planning year, and used the administrative regions of Changge City, Yuzhou City, Weidu District, Xuchang County and Yanling County as the computing units in the first-tier allocation. Water users were divided into urban domestic water consumption, rural domestic water consumption, agricultural water consumption, industrial water consumption, tertiary industrial water consumption, the eco-environmental water demand of the river system and the external eco-environmental water. The water supply sources included the basic water supply (local surface water, groundwater, transit water, water diverted from the Yellow River and the South-to-North Water Transfer Project) and the unconventional water sources (the effluent from sewage treatment plants).

The tributary inlets and sections monitored for water quality, defined as control nodes for the optimal distribution of river water resources, were generalized in each administrative area so that the scale could be transformed between the computing units of the first tier and the control nodes of the second and third tiers. The partition results of the control nodes included Luma Bridge, Gaocun Bridge, Nanwaihuan Bridge, Sanzhang Gate, Dashi Bridge and Local Railway Bridge (Figure 2).

For the eco-environmental water requirements of the Qingyi River, we used the values calculated by the Major Science and Technology Program for Water Pollution Control and Treatment (Yu et al. 2016), as listed in Table 1. The minimum monthly average flow method and the wetted perimeter method are used to calculate the eco-environmental water requirements.

Table 1

Target of eco-environmental water requirements of the Qingyi River, Xuchang City (m3/s)

River control nodesLuma BridgeGaocun BridgeNanwaihuan BridgeSanzhang GateDashi BridgeLocal Railway Bridge
Jan 0.9 1.9 0.2 0.2 0.4 0.2 
Feb 0.9 1.9 0.2 0.2 0.4 0.2 
Mar 0.4 0.3 0.2 0.4 0.2 
Apr 0.4 1.1 0.3 0.2 0.4 0.3 
May 0.4 1.1 0.3 0.3 0.4 0.3 
Jun 0.4 1.2 0.3 0.3 0.4 0.4 
Jul 1.3 3.3 0.7 0.8 1.1 
Aug 1.3 3.2 0.7 0.7 
Sep 1.3 2.7 0.7 0.5 0.8 0.6 
Oct 1.3 2.7 0.7 0.2 0.8 0.3 
Nov 0.9 1.9 0.2 0.2 0.3 0.2 
Dec 0.9 1.9 0.2 0.2 0.3 0.2 
River control nodesLuma BridgeGaocun BridgeNanwaihuan BridgeSanzhang GateDashi BridgeLocal Railway Bridge
Jan 0.9 1.9 0.2 0.2 0.4 0.2 
Feb 0.9 1.9 0.2 0.2 0.4 0.2 
Mar 0.4 0.3 0.2 0.4 0.2 
Apr 0.4 1.1 0.3 0.2 0.4 0.3 
May 0.4 1.1 0.3 0.3 0.4 0.3 
Jun 0.4 1.2 0.3 0.3 0.4 0.4 
Jul 1.3 3.3 0.7 0.8 1.1 
Aug 1.3 3.2 0.7 0.7 
Sep 1.3 2.7 0.7 0.5 0.8 0.6 
Oct 1.3 2.7 0.7 0.2 0.8 0.3 
Nov 0.9 1.9 0.2 0.2 0.3 0.2 
Dec 0.9 1.9 0.2 0.2 0.3 0.2 

Analysis of the water quality monitoring data from 2010 to 2011 from five sections of the main Qingyi River showed that the main pollutants were COD and ammonia nitrogen (Chang et al. 2015). We set the water quality target for the Qingyi River as Class IV of the Surface Water Quality Standards, outlined in the Water Environment Function Blocks and Water Function Regionalization of Henan Province. Therefore, the water quality target was to achieve class IV concentrations of the main pollutants, COD and ammonia nitrogen.

Effluent from sewage treatment plants belongs to standard A for pollutants discharged from municipal wastewater treatment plants, and other water sources belong to Class III of the surface water quality standards.

Establishment of allocation scenarios

For the water quantity allocation scenario, the low save-water scenario is the scheme under the current water-saving level. The high save-water scenario of the water resources current situation is the scheme to reduce the water demand of planning years by limiting the development of the high water consuming industries, promoting the water saving instrument and engineering, and carrying out the infrastructure renovation of water storage, water supply, water use, drainage and other aspects.

For the water quality allocation scenario, parameters adjustment is to control R of river ecological environment water supply source in the first-tier macro allocation. In this scheme, pollution reduction measures refer in particular to Wetland construction, as it was selected in Qingyi River. Partial effluent from sewage treatment plants was discharged into the wetland project for advanced treatment so as to reduce the amount of pollution into the river. The allocation schemes are listed in Table 2.

Table 2

Allocation scenarios for the Qingyi River, Xuchang City

Scenario nameScenario 1Scenario 2Scenario 3Scenario 4Scenario 5
Water quantity allocation scenario The low save-water scenario √     
The high save-water scenario  √ √ √ √ 
Water quality allocation scenario Ratio of wastewater to clean water   √  √ 
Wetland construction    √ √ 
Scenario nameScenario 1Scenario 2Scenario 3Scenario 4Scenario 5
Water quantity allocation scenario The low save-water scenario √     
The high save-water scenario  √ √ √ √ 
Water quality allocation scenario Ratio of wastewater to clean water   √  √ 
Wetland construction    √ √ 

Calibration of water quality model parameters

When calibrating the water quality model, we used the relevant results of the Major Science and Technology Program for Water Pollution Control and Treatment (Zhang et al. 2015), namely a diffusion coefficient of 0.15 m2/s for the Qingyi River, and ammonia nitrogen and COD degradation coefficients of 0.18/d and 0.15/d, respectively.

RESULTS AND DISCUSSION

First-tier optimal allocation of water resources in the river basin

When COD was the control factor, R = 1/1; when ammonia nitrogen was the control factor, R = 1/7. Due to lack of natural runoff, the Qingyi River received industrial and domestic wastewater along its entire course, it belonged to the low-flow polluted river. If R = 1/7, a large amount of clean water was needed, it was not in line with the characteristics of Qingyi River, so we used R = 1/1.

We calculated the eco-environmental water requirement of the river by converting the instantaneous flow of each node into the annual water demand, calculated the operation model for the time of the year, and obtained the results for the different scenarios for a water supply guarantee rate of 50% in 2020. The results of the configuration of the various schemes are in line with the requirements of Implementation Plan of Water Ecological Civilization City Construction of Xuchang City, and the water supply guarantee rate of water users. The guaranteed rate of water consumption of urban life, rural life, agriculture, industry, tertiary industry, the eco-environmental water demand of the river system, and the external eco-environmental water were designed to 95%, 95%, 75%, 95%, 95%, 80% and 70%, respectively. Configuration results of river eco-environmental water demand in the first-tier macro allocation were analyzed to recommend feasible schemes for the second-tier macro allocation model, as shown in Table 3.

Table 3

First-tier optimal allocation of water resources (104 m3)

Scenario nameAllocated water of river ecological environmentWeidu DistrictXuchang CountyChangge CityYuzhou CityTotal
Scenario 1 Basic water supply 1,702 1,115 337 3,154 
Effluent from sewage treatment plants 926 1,497 803 3,226 
Water shortage 196 796 163 1,156 
Water supply guarantee rates 100 94 71 67 85 
Scenario 2 Basic water supply 1,785 1,397 375 3,557 
Effluent from sewage treatment plants 926 1,591 803 3,320 
Water shortage 19 514 125 659 
Water supply guarantee rates 100 99 81 75 91 
Scenario 3 Basic water supply 463 1,804 1,406 375 4,048 
Effluent from sewage treatment plants 463 1,591 746 2,800 
Water shortage 562 125 688 
Water supply guarantee rates 100 100 79 75 91 
Scenario 4 Basic water supply 1,785 1,397 375 3,557 
Effluent from sewage treatment plants 463 796 402 1,661 
Wetland effluent 463 796 402 1,661 
Water shortage 18 513 125 657 
Water supply guarantee rates 100 99 81 75 91 
Scenario 5 Basic water supply 463 1,804 1,406 375 4,048 
Effluent from sewage treatment plants 232 796 373 1,401 
Wetland effluent 232 796 373 1,401 
Water shortage 562 125 688 
Water supply guarantee rates 100 100 79 75 91 
Scenario nameAllocated water of river ecological environmentWeidu DistrictXuchang CountyChangge CityYuzhou CityTotal
Scenario 1 Basic water supply 1,702 1,115 337 3,154 
Effluent from sewage treatment plants 926 1,497 803 3,226 
Water shortage 196 796 163 1,156 
Water supply guarantee rates 100 94 71 67 85 
Scenario 2 Basic water supply 1,785 1,397 375 3,557 
Effluent from sewage treatment plants 926 1,591 803 3,320 
Water shortage 19 514 125 659 
Water supply guarantee rates 100 99 81 75 91 
Scenario 3 Basic water supply 463 1,804 1,406 375 4,048 
Effluent from sewage treatment plants 463 1,591 746 2,800 
Water shortage 562 125 688 
Water supply guarantee rates 100 100 79 75 91 
Scenario 4 Basic water supply 1,785 1,397 375 3,557 
Effluent from sewage treatment plants 463 796 402 1,661 
Wetland effluent 463 796 402 1,661 
Water shortage 18 513 125 657 
Water supply guarantee rates 100 99 81 75 91 
Scenario 5 Basic water supply 463 1,804 1,406 375 4,048 
Effluent from sewage treatment plants 232 796 373 1,401 
Wetland effluent 232 796 373 1,401 
Water shortage 562 125 688 
Water supply guarantee rates 100 100 79 75 91 

Because of base flow scarcity in Qingyi River, runoff was neglected in the first-tier model. As shown in Table 3, the eco-environmental water requirement guarantee rate of scenario 1 was 85%. Under enhanced water-saving conditions, scenarios 2, 3, 4, and 5 predicted water supply guarantee rates greater than 90%. There were no constraint conditions of R for scenarios 1 and 2. The water source of the river in Xuchang County and Changge City include clean water and sewage, and clean water is the only source in Yuzhou City; however, the water pollution status was predicted to worsen as sewage was the only source of water to the river in the Weidu District. Scenarios 3 and 5 had R; scenario 4 had no constraint condition of R. However, this scenario can achieve the aim of improved water quality with wetlands to reduce the pollutant loads into the river. From these results, we recommend scenarios 3, 4, and 5.

Second tier of the optimal river water quantity allocation model

We derived the optimal river water allocations of scenarios 3, 4, and 5, using the water allocated to the river ecological environment in the various administrative regions from the first tier as the water supply and the operation model calculated for each month, and we have listed the eco-environmental water requirement guarantee rate of each control node in each month (Tables 4 and 5).

Table 4

Eco-environmental water requirement guarantee rate of scenarios 3 and 5 (%)

Control nodeJanFebMarAprMayJunJulAugSepOctNovDec
Luma Bridge 92 100 100 100 100 100 63 63 65 63 95 92 
Nanwaihuan Bridge 100 100 88 80 80 80 87 87 85 81 100 100 
Sanzhang Gate 83 83 100 88 81 83 80 80 82 81 83 83 
Local Railway Bridge 100 100 100 96 88 81 80 82 80 100 100 100 
Dashi Bridge 100 93 93 93 93 89 88 86 87 82 100 100 
Gaocun Bridge 92 86 89 89 87 86 81 83 80 83 83 82 
Control nodeJanFebMarAprMayJunJulAugSepOctNovDec
Luma Bridge 92 100 100 100 100 100 63 63 65 63 95 92 
Nanwaihuan Bridge 100 100 88 80 80 80 87 87 85 81 100 100 
Sanzhang Gate 83 83 100 88 81 83 80 80 82 81 83 83 
Local Railway Bridge 100 100 100 96 88 81 80 82 80 100 100 100 
Dashi Bridge 100 93 93 93 93 89 88 86 87 82 100 100 
Gaocun Bridge 92 86 89 89 87 86 81 83 80 83 83 82 
Table 5

Eco-environmental water requirement guarantee rate of scenario 4 (%)

Control nodeJanFebMarAprMayJunJulAugSepOctNovDec
Luma Bridge 95 100 100 100 100 100 65 65 67 65 98 95 
Nanwaihuan Bridge 100 100 88 80 80 80 87 87 85 82 80 80 
Sanzhang Gate 82 82 82 81 82 82 81 64 80 82 80 83 
Local Railway Bridge 100 100 100 100 100 100 100 100 100 100 100 100 
Dashi Bridge 100 100 93 93 93 89 88 86 87 81 100 100 
Gaocun Bridge 89 91 89 97 87 86 80 80 82 82 83 82 
Control nodeJanFebMarAprMayJunJulAugSepOctNovDec
Luma Bridge 95 100 100 100 100 100 65 65 67 65 98 95 
Nanwaihuan Bridge 100 100 88 80 80 80 87 87 85 82 80 80 
Sanzhang Gate 82 82 82 81 82 82 81 64 80 82 80 83 
Local Railway Bridge 100 100 100 100 100 100 100 100 100 100 100 100 
Dashi Bridge 100 100 93 93 93 89 88 86 87 81 100 100 
Gaocun Bridge 89 91 89 97 87 86 80 80 82 82 83 82 

As shown in Tables 4 and 5, under enhanced water-saving and R = 1/1, the eco-environmental water requirement guarantee rate of each control node in each month was more than 80% except for the Luma Bridge section in scheme 3. Under enhanced water-saving and using wetland, the guarantee rate of scenario 4 was more than 80% except for the Luma Bridge and Sanzhang Gate section. Under enhanced water-saving, R = 1/1 and wetlands added, the guarantee rate of scenario 5 was more than 80% except for the Luma Bridge section. There was no significant difference between the three scenarios, so we simulated the water quality from the results of scenarios 3, 4, and 5 from the second tier.

Water quality simulation results from the third tier of the model

We used the MIKE model to obtain the results for the river water quality for scenarios 3, 4, and 5. The water quality target achievement rates for each control node are shown in Figures 3 and 4.

Figure 3

COD water quality target achieved at each control node.

Figure 3

COD water quality target achieved at each control node.

Figure 4

Ammonia nitrogen water quality target achieved at each control node.

Figure 4

Ammonia nitrogen water quality target achieved at each control node.

Figures 3 and 4 show that under scenario 3, the COD concentration target at the Luma Bridge section was achieved less than 100% of the time and the ammonia nitrogen concentration target was achieved less than 100% of the time in the Luma Bridge, Nanwaihuan Bridge, Dashi Bridge, and Gaocun Bridge sections. For scenario 4, the COD was achieved less than 100% of the time in the Luma Bridge section, and the ammonia nitrogen was achieved less than 100% of the time in the Luma Bridge and Gaocun Bridge sections. For scenario 5, the COD was achieved less than 100% of the time in the Luma Bridge section, and the ammonia nitrogen was achieved less than 100% of the time in the Luma Bridge and Gaocun Bridge sections. Overall, the compliance was highest for scenario 5, followed by scenario 4, and was lowest for scenario 3. We therefore recommended scenario 5.

Optimal allocation of river water resources for the recommended scenario

The water allocation results and the water quality simulation for each control node, from an analysis of the operation results of the five scenarios in the three-tier model, are shown in Table 4 and in Figures 5 and 6.

Figure 5

Simulation of the monthly COD concentrations of the recommended scenario.

Figure 5

Simulation of the monthly COD concentrations of the recommended scenario.

Figure 6

Simulation of the monthly ammonia nitrogen concentrations of the recommended scenario.

Figure 6

Simulation of the monthly ammonia nitrogen concentrations of the recommended scenario.

Table 4 shows that the overall water quantity in the recommended scenario was good and the eco-environmental water requirement guarantee rate of each control node in each month was more than 80%, except for the Luma Bridge section. Serious shortage of water was predicted in the Luma Bridge section for July, August, September and October, and the water shortage rates were 63%, 63%, 65% and 63%, respectively. We recommend that the technology on rainwater collection, storage and treatment showed be developed by the government, aimed at increasing the water supply source for the river ecological environment.

As shown in Figures 5 and 6, the model predicted good overall water quality for the recommended scenario, with overall ammonia nitrogen and COD compliance rates of up to 82% and 96%, respectively. The model predicted that the ammonia nitrogen concentration would not meet the standards in individual months at the Luma Bridge, Nanwaihuan Bridge, and Gaocun Bridge sections, and that COD concentration would not meet the standard in individual months at the Luma Bridge and Gaocun Bridge sections. In the recommended scenario, sewage is the main supply at substandard nodes. We recommend that Xuchang City shut down a number of high pollution emission enterprises and increase the scale of wetland construction to improve river water quality.

CONCLUSIONS

To solve river eco-environmental problems, we proposed a three-tier model to determine the optimal allocation of the river water quantity and the river water quality. The output of the previous level is the input of the next level. We established a set of water quantity and quality allocation scenarios. This approach provides a new method for allocating water resources based on river water quality and water quantity.

Using the Qingyi River (Xuchang section) as an example, we found a solution that predicted an overall good status for both the water quantity and water quality, with more than 80% of the eco-environmental water requirement achieved at each node per month except for the Luma Bridge section, and overall ammonia nitrogen and COD concentration compliance rates of 82% and 96%, respectively. The ammonia nitrogen concentration did not meet the standards in individual months at the Luma Bridge, Nanwaihuan Bridge, and Gaocun Bridge sections, and the COD concentrations were below the standard in individual months at the Luma Bridge and Gaocun Bridge sections. The water quality in the remaining sections can be improved from the current class V to class IV with the proposed scenario. The sections the pollution from the sections where the standards are not met can be reduced by enhancing the water treatment and increasing wetland development. We used this example to test the feasibility of the three-tier model in obtaining the optimal distribution of river water resources.

ACKNOWLEDGEMENTS

The authors thank Major Science and Technology Program for Water Pollution Control and Treatment (No. 2015ZX07204-002) and the National Natural Science Foundation of China (No. 51379191) for their partial financial support that made this project possible.

REFERENCES

REFERENCES
Avogadro
,
E.
,
Minciardi
,
R.
&
Paolucci
,
M.
1997
A decisional procedure for water resources planning taking into account water quality constrains
.
European Journal of Operational Research
102
,
320
334
.
Carlos
,
P.
,
Gidon
,
O.
&
Abraham
,
M.
1997
Optimal operation of regional system with diverse water quality sources
.
Journal of Water Resources Planning and Management
123
,
105
115
.
Chang
,
S.
,
Liu
,
Y.
,
Fu
,
Q.
&
Yang
,
G.
2015
Water pollution cause analysis and control strategies of Xuchang Section of Qingyi River
.
Journal of Environmental Engineering Technology
6
,
471
477
.
Gao
,
F.
,
Huang
,
Q.
&
Zhang
,
H. B.
2012
Water resources regulation in the Baoji reach of Weihe River based on the protection of river ecological base flow
.
Journal of Arid Land Resources and Environment
6
,
149
154
.
Hatry
,
C.
,
Binder
,
T. R.
,
Thiem
,
J. D.
,
Hasler
,
C. T.
,
Smokorowski
,
K. E.
,
Clarke
,
K. D.
,
Katopodis
,
C.
&
Cooke
,
S. J.
2013
The status of fishways in Canada: trends identified using the national CanFishPass database
.
Reviews in Fish Biology and Fisheries
23
(
3
),
271
281
.
Hu
,
S.
2012
The Study on Water Resources Optimal Allocation of City Agglomeration with Balanced Development
.
Master's Thesis
,
Zhengzhou University
,
Zhengzhou
,
China
.
Liu
,
C. M.
2002
The concept and importance of ecological water demand
.
The Impact of Science on Society
2
,
25
29
.
Meng
,
Y.
,
Zhang
,
X.
,
Xia
,
J.
,
Wu
,
S. F.
,
Wang
,
J. C.
&
Christopher
,
J. G.
2016
Definition of environmental flow components for Leiocassis Longirostris in the Huai River considering habitat change and hydrological change
.
Journal of Hydraulic Engineering
5
,
626
634
.
Penas
,
F. J.
,
Juanes
,
J. A.
&
Álvarez-Cabria
,
M.
2014
Integration of hydrological and habitat simulation methods to define minimum environmental flows at the basin scale
.
Water and Environment Journal
28
(
2
),
252
260
.
Peng
,
S. M.
,
Zheng
,
S. K.
,
Wang
,
Y.
,
He
,
L. P.
,
Liu
,
J.
&
Wang
,
W. H.
2016
Study on integrated allocation and dispatch model of water quality and quantity for the Yellow River
.
Advances in Water Science
2
,
196
205
.
Pingry
,
D. E.
,
Shaftel
,
T. L.
&
Boles
,
K. E.
1990
Role for decision-support systems in water-delivery design
.
Journal of Water Resources Planning and Management
116
,
629
644
.
Song
,
J. X.
,
Li
,
H. N.
&
Wang
,
B. Z.
2003
Summary on research of ecological and environmental water requirement of river
.
Journal of Soil and Water Conservation
17
,
95
97
.
Wang
,
J. H.
,
Hou
,
B. D.
,
Jiang
,
D. C.
&
Xiao
,
W. H.
2016
Optimal allocation of water resources based on water supply security
.
Water
8
,
237
.
Yu
,
L. J.
,
Yang
,
Z.
&
Wang
,
H. Y.
2016
A Research Report on key Technologies Integration and Comprehensive Demonstration of Water Environment Quality Improvement and Functional Recovery in Qingyi River Basin
.
Zhengzhou University
,
Zhengzhou
,
China
.
Zhang
,
W.
,
Wang
,
Y.
,
Peng
,
H.
,
Li
,
Y.
,
Tang
,
J.
&
Wu
,
K. B.
2010
A coupled water quantity–quality model for water allocation analysis
.
Water Resources Management
24
,
485
511
.
Zhang
,
H.
,
Feng
,
M. Q.
,
Wang
,
L.
&
Nie
,
W. S.
2015
Research on water environmental effects of Qingyi River Based on ecological water diversion
.
Journal of Xi'an University of Technology
1
,
83
90
.