Research on Joint Regulation Strategy of Water Conservancy Project Group in the Multi-Branch Channels of the Ganjiang River Tail for Coping with Dry Events

Abstract

The problem of low water level and uneven distribution of flow in the multi-branch channels at the tail of the Ganjiang River (GJRT) during the dry season has been affecting the local water supply, navigation, and aquatic ecological environment. In recent years, water conservancy projects have been built in each branch of the multi-branch channels at the GJRT. Finding a way to utilize the water conservancy project group to carry out joint regulation and meet the water level and discharge requirements of each branch is an important issue that urgently needs to be solved. This paper analyzes the hydrodynamic process and its impact on water supply, navigation, and ecology in multi-branch channels without water conservation projects through hydrological data analysis and numerical simulation. By conducting numerical experiments on joint regulation of water conservation project group, a multi-objective regulation strategy is proposed to meet the water level and discharge of each branch. The results indicate that the discharge at the GJRT has been continuously decreasing from 1 September. Due to the jacking effect of Poyang Lake, the water level plunges at the GJRT from 1 October, which occurred later than the decrease in water level. The disruption of water levels and discharge makes it difficult to meet the regional water demand. The optimal time to initiate regulation is 1 October, and the target water level of Waizhou Station is 15.5 m, located upstream of the Ganjiang River tail. When the water level before each branch project gate is uniform and exceeds 15.5 m, the water level of Waizhou Station satisfies the requirement. However, the discharge of each branch does not meet the demand. In contrast to a scheduling regulation strategy that maintains the same water level in front of each gate, adopting a strategy with different water levels before each gate can effectively adjust the diversion ratio and fulfill the discharge demand of each branch at the tail of the Ganjiang River.

1. Introduction

With the progression of climate change and the intensification of human activities, the hydrodynamic processes in rivers are undergoing continuous alterations. This phenomenon has given rise to a series of challenges concerning water supply, navigation, and the regional ecological environment [1]. For instance, during the severe drought that occurred in the Poyang Lake Delta of China, the concentrations of water quality parameters (i.e., TN and TP) rose by 50.2% and 240%, respectively [2]. Due to the lack of water supply, the poor hydrodynamic conditions in Yilong Lake, China, made its water quality not meet the demand [3]. Against the backdrop of drought in Poyang Lake, the long-term downward trend in water levels may give rise to potential navigation safety hazards [4]. The water level variation significantly impacted the wetland vegetation (p < 0.01) in the Poyang Lake Basin, and the vegetation greenness declined under the influence of increasing drought in recent years [5]. The periodic droughts intensified the degradation of aquatic habitats in the U.S. Praia wetlands and reduced the number of migratory birds [6].

Regulating the hydrodynamic conditions of rivers through water conservancy projects can improve water level distribution and flow allocation, thereby influencing the overall hydrodynamic processes of the river basin [7]. Therefore, research on river regulation and its positive effects has received increasing interest. Li et al. [8] considered the effects of upstream reservoir operation on cyprinid fish habitat quality in the Lijiang River, proposing an ecologically based flow regime for fish conservation. Cao et al. [9] comprehensively evaluated the economic and ecological benefits of the water conservancy project located in Daling River, China, proposing an ecological operation scheme for projects to increase the ecological release. Chen et al. [10] pointed out that the regulation of water conservancy projects could directly change hydrodynamics and mitigate wetland vegetation deterioration. Xiao et al. [11] found that the implementation of water conservancy projects could expand the inundation area of wetland upstream and improve the vegetation species composition and community structure.

With the further development of river regulation, cascade hydraulic projects have gradually emerged along rivers. In recent years, numerous studies have been conducted on the joint regulation strategy of cascade water conservancy projects. Xu et al. [12] put forward long-term ecological operation scenarios for the cascade reservoirs along the Jinshan River and these scenarios have exerted positive effects on the river ecosystem. Xiao et al. [13] increased the water depth and velocity in the river section downstream of cascade projects along the Yalong River, by regulating the cascade projects with suitable ecological flow interval. Most existing studies focus on the regulating strategy of a single water conservancy project or cascaded water conservancy projects on an individual river. However, the river structure in plain river network areas is more complex with many projects, and there is still deficiency in the research on joint regulation strategy for river network water conservancy groups.

The GJRT is characterized by the confluence of the river and the lake. In this area, the water level changes and flow distribution in the multi-branch channels are influenced by the backwater effect of Poyang Lake. Poyang Lake is the largest freshwater lake in China, with its northern part connected to the Yangtze River. The complex interactions between the Yangtze River and Poyang Lake create a “dry-rise-flood-recession” hydrological regime of Poyang Lake [14]. Under this hydrological regime, the lake exhibits significant intra-annual water level fluctuations, which makes the hydrodynamic process in the GJRT more complex. In the 21st century, the water level of Poyang Lake has shown a declining trend, frequently dropping below the historical lowest water level during the dry season [15]. The drought in Poyang Lake has triggered a series of ecological and environmental issues in the lakeside wetlands. For instance, the accelerated water recession of Poyang Lake caused the early exposure of wetland vegetation, making it difficult for roots to accumulate sufficient nutrients [16].

To cope with the drought conditions, a series of water conservancy projects have been built at the GJRT, forming a relatively complete water conservancy system. How to find joint regulation strategy of water conservancy project group in the multi-branch channels of the GJRT is of significance for meeting the water demand for water supply, navigation, and ecological environment in the GJRT during the dry season. Compared with previous studies on water conservancy project regulation mentioned above, there are primarily two challenges in the regulation of water conservancy projects in the GJRT: (1) as the GJRT is connected to Poyang Lake, the regulation strategy must consider the jacking effect of Poyang Lake; (2) given that the GJRT is a multi-branch channel, the regulation strategy needs to consider the mutual influences among different tributaries.

The novelty of this study is characterized by proposing a multi-objective regulating strategy for multi-branch river system connected to the lake with significant annual water level fluctuations. This study explores the following two main aspects: (1) analyzing the hydrodynamic processes of the GJRT under the interaction between river and lake, thereby identifying the proper water storage timing to achieve the full utilization of water resources at the end of flood season; (2) considering the mutual discharge response relationships among different branches in the muti-branch channels, proposing joint regulating strategies for water conservancy project group that simultaneously meets the discharge demand of each branch.

2. Materials and Methods

2.1. Study Area

The tail of the Ganjiang River (GJRT) is located on the southern bank of the middle and lower reaches of the Yangtze River in China, covering Nanchang City and Jiujiang City of Jiangxi Province (Figure 1a). It serves as a key transitional zone before the Ganjiang River flows into Poyang Lake. As the largest river in the Poyang Lake Basin, the Ganjiang River has a total length of 823 km and a drainage area of 82,809 km² (accounting for 50% of the total area of Jiangxi Province), which contributes more than 50% of the water volume of Poyang Lake [17].

The GJRT features a typical alluvial plain, while it has a multi-branch channels structure. Starting from Waizhou Station, the Ganjiang River is divided into four branches, namely the main branch, north branch, middle branch, and south branch. Waizhou Station is the last hydrological station on the Ganjiang River before it flows into Poyang Lake. Therefore, it is generally regarded as the control station for the GJRT. The water conservancy project group in the GJRT consists of the main branch project, north branch project, middle branch project, and south branch project, which are distributed in the four branches (Figure 1b). Different from natural multi-stage braided rivers, the GJRT is constrained by embankments along all its branches, as Nanchang City—where the GJRT is situated—serves as the provincial capital of Jiangxi Province with a dense population and well-developed agriculture. Consequently, the regulation of the water conservancy projects will not significantly alter the river channel morphology in this region.

2.2. Two-Dimensional Hydrodynamic Model of the GJRT

2.2.1. Model Establishment

A two-dimensional hydrodynamic model of the GJRT was established to simulate the hydrologic and hydrodynamic processes of each branch in the GJRT. The model was established based on 2014 DEM of the GJRT (Figure 2a) and based on Mike21 (Danish Hydraulic Institute Co., Ltd., Shanghai, China) developed by the Danish Hydraulic Institute (DHI). The governing equation of the model is the two-dimensional shallow water equation, which is derived by integrating the momentum and mass conservation equations vertically. The two-dimensional shallow water equation calculates the vertical average flow velocity, making it suitable for simulating the flow fields of shallow areas, such as lakes and rivers.

Considering the computational efficiency of the model and the complexity of bed morphology, different types and sizes of meshes were applied to the main channel and beach area at the GJRT (Figure 2b). For the beach with flatbed morphology, the quadrilateral mesh with a side length of 50–100 m was adopted. For the main channel with undulating bed morphology, the quadrilateral mesh with a side length of 100–200 m was used. Additionally, triangular meshes were employed between the beach area and the main channel.

2.2.2. Calibration and Verification

(1)

Calibration

The setting of Manning roughness coefficients for each branch was mainly based on the studies of Zong et al. [18] and Peng et al. [19].

Table 1. The comparison of Manning roughness coefficients setting for the GJRT in different references.

	Main Branch	North Branch	Middle Branch	South Branch
Zong et al. [18]	0.018	0.033	0.028~0.03	0.026
Peng et al. [19]	0.025	0.032~0.033	0.028~0.03	0.026–0.027
This paper	0.016~0.035	0.021~0.033	0.011~0.032	0.015~0.033

presents a comparison of parameter settings between this paper and previous studies. The Manning roughness coefficient values used in this study are largely consistent with those reported in previous studies. The key modification lies in the slight adjustments made to the coefficients for the main channels and floodplains of each branch, which were derived from the parameter calibration results of this study. The water level data measured downstream of Waizhou Station on 26 May 2014 (when the discharge of Waizhou Station was 9830 m³/s) was used for calibration, which indicates that 91% of the simulated values had errors less than 0.2 m, and the coefficient of determination (R²) was 0.92. The bias, RMSE, and NSE were 0.07, 0.14, and 0.91, respectively (see Figure 3).

(2)

Verification

To further demonstrate the model’s robustness, we conducted additional simulations of the hydrodynamic processes at the GJRT during 2023. The inflow discharge of the model, which is the discharge of Waizhou Station in 2023, is shown in Figure 4. The bias, RMSE, and NSE of Changyi (main branch), Jiangbu (north branch), Louqiain (middle branch), and Chucha (south branch) are listed in

Table 2. The bias, RMSE, NSE of simulated results for each branch.

	Main Branch	North Branch	Middle Branch	South Branch
Bias	−0.55	−0.02	0.15	0.54
RMSE	0.63	0.41	0.44	0.67
NSE	0.90	0.72	0.93	0.89

. A comparison of the simulated and observed water level of four tributaries in 2023 is shown in Figure 5. The validation results indicate that the model can accurately simulate the hydrodynamic conditions of different tributaries at the GJRT under varying inflow discharges throughout the year.

2.3. Design Framework for Regulation Scenarios

To comprehensively consider the jacking effect of Poyang Lake and the mutual influences among different tributaries, the design steps for the water control project regulation scenarios at the GJRT are as follows:

(1)

Analysis of hydrological conditions at the GJRT under the influence of Poyang Lake water level and the Ganjiang River runoff.

(2)

Based on the principle of fully utilizing the end-of-flood-season runoff in the Ganjiang River, determining the regulation timing for the water conservancy project at the GJRT.

(3)

Analyzing the water level and discharge demands of each tributary, determining the regulation objectives for the water conservancy project group at the GJRT.

(4)

Considering the interactions among different tributaries, conducting numerical regulation experiments to identify reasonable regulation scenarios.

3. Results

3.1. Analysis of Hydrological Situation in Flood–Dry Transition Period at the GJRT

To effectively utilize water resources at the end of the flood season and cope with the dry event, the transition stage between flood season and dry season is a critical phase for regulating the water conservation project group. Given that the flood season at the GJRT is from April to June, the flood–dry transition period is from May to October.

(1)

Water level

The water level variation process from May to October is shown in Figure 6. From 1 May to 31 August, the water level of Waizhou Station experienced several significant increases, reaching a peak water level of 21 m. May and June were the flood season for the Ganjiang River, while July and August were the flood season of Poyang Lake. The alternating occurrence of floods in the Ganjiang River and Poyang Lake led to significant fluctuations in the water level at the GJRT.

From 1 to 30 September, the water level at Waizhou Station remained stable at around 17.2 m, which is associated with the jacking effect of Poyang Lake. As the retreating process of Poyang Lake has not yet ended in September, the jacking of Poyang Lake on the GJRT hindered the retreat of this area. From 1 to 31 October, the GJRT entered the retreating stage, with a sharp drop in water level. During the flood–dry transition period, the water level process of each branch of the GJRT was basically consistent with that of Waizhou Station. The water level dropping in the GJRT would influence the growth of wetland vegetation. Tan [20] revealed that the grassland area in the southern part of Poyang Lake (e.g., the GJRT) exhibited greater sensitivity to fluctuations in water levels compared to other regions within the lake basin.

(2)

Discharge

The discharge varying process from May to October was shown in Figure 7a. Before 1 September, the water level and discharge showed synchronous changes. After 1 September, the discharge of Waizhou Station gradually decreased. Unlike the trend of discharge, the water level at Waizhou Station began to decrease on 1 October. After September, the inconsistency in the trend of water level and discharge changes at Waizhou Station is mainly due to the jacking effect of Poyang Lake.

The discharge of each branch exhibited the same trend as that of Waizhou Station. Specifically, the discharge of the south branch turned negative after July, indicating that the jacking effect of Poyang Lake in flood season led to the backflow of lake water into the south branch (Figure 7b).

3.2. The Joint Regulation Strategy of Water Conservation Hub Group at the GJRT

(1)

Regulation Objectives

According to the “GJRT Improvement Plan”, the regulation objectives include ensuring the overall water level of the GJRT and discharge of each branch in the braided river channel. The water level requirement for the GJRT is that the water level of Waizhou Station is not less than 15.5 m. The discharge demand for the GJRT is as follows:

(1). When the discharge of Waizhou Station is 488.7 m³/s, it could meet the ecological, production and living, and navigation demands during the dry season.

(2). When the discharge of the main branch is 341 m³/s, it could meet the ecological and navigation demands during the dry season.

(3). When the discharge of the northern branch is 5.1 m³/s, it could meet the ecological water demand during the dry season.

(4). When the discharge of the middle branch is 60.4 m³/s, it could meet the ecological and production and living water demands during the dry season.

(5). When the discharge of the southern branch is 61 m³/s, it could meet the ecological and navigation demands.

(2)

Regulation Scenarios

The principle of the regulation scenario is to fully utilize the incoming water during flood–dry transition period to ensure the water level and discharge demand during the dry season. The water level decreases rapidly after 1 October (Figure 6a), which makes water storage more challenging. The starting time for the hub regulation should be set on 1 October, as the discharge of Waizhou Station was 800 m³/s. The regulating timing is associated with the influence of Three Gorges Dam (TGA) on Poyang Lake water level. After the operation of the TGA in 2003, the relationship between the Yangtze River and Poyang Lake underwent significant changes. Due to the impoundment of TGA at the end of the flood season, Poyang Lake water level in October–December decreased [21].

Setting the input discharge as 800 m³/s, we conducted the regulation numerical experiment by adjusting the water level before the gate of each project. To satisfy the overall water level demand (15.5 m), we synchronously set the water levels before each gate to 13.5, 14.5, 15.5, and 16.5 m for RS-1 to RS-4, respectively (

Table 3. Equal water level scenarios.

	Main Branch	North Branch	Middle Branch	South Branch
RS-1	13.5	13.5	13.5	13.5
RS-2	14.5	14.5	14.5	14.5
RS-3	15.5	15.5	15.5	15.5
RS-4	16.5	16.5	16.5	16.5

).

Table 4. Regulation effect comparison for equal water level scenarios.

RS-1	Waizhou	Main	North	Middle	South
Water level (m)	13.8	13.5	13.5	13.5	13.5
Demand discharge (m3/s)	488.7	341	5.1	60.4	61
Actual discharge (m3/s)	800	527.77	35.6	108.26	128.37
RS-2	Waizhou	Main	North	Middle	South
Water level (m)	14.6	14.5	14.5	14.5	14.5
Demand discharge (m3/s)	488.7	341	5.1	60.4	61
Actual discharge (m3/s)	800	375.7	89.22	161.16	173.92
RS-3	Waizhou	Main	North	Middle	South
Water level (m)	15.5	15.5	15.5	15.5	15.5
Demand discharge (m3/s)	488.7	341	5.1	60.4	61
Actual discharge (m3/s)	800	324.27	128.87	174.32	172.54
RS-4	Waizhou	Main	North	Middle	South
Water level (m)	16.5	16.5	16.5	16.5	16.5
Demand discharge (m3/s)	488.7	341	5.1	60.4	61
Actual discharge (m3/s)	800	289.82	159.15	195.97	155.06

shows the regulation effect for equal water level scenarios. For RS-1 and RS-2, under the condition of the water level in front of each gate lower than 15.5 m, the water level at Waizhou Station could not reach the regulation target. The discharge of the main branch was 324.27 m³/s for RS-3, which decreases to 289.82 m³/s for RS-4. For RS-3 and RS-4, raising the water level before each gate could effectively improve the water level of Waizhou Station, but it could not meet the discharge demand of the main branch. The evaluation of regulation scenarios (RS1-RS-4) is shown in

Table 5. Evaluation of regulation scenarios (RS-1–RS-4).

Demand	RS-1	RS-2	RS-3	RS-4
Water level	×	×	✓	✓
Discharge	✓	✓	×	×

. The result shows that adopting the equal water level scenarios would face the problem of unreasonable discharge distribution in the multi-level branching river in the GJRT.

To increase the discharge of the main branch while ensuring the water level demand of Waizhou Station, this paper proposes different water level scenarios for water levels before each gate (

Table 6. Difference water level scenarios.

	Main Branch	North Branch	Middle Branch	South Branch
RS-5	15.5	15.5	15.5	15.6
RS-6	15.5	15.5	15.6	15.5
RS-7	15.5	15.6	15.5	15.5
RS-8	15.4	15.5	15.5	15.5

). The principle of the scenario is to lower the water level in the main branch (RS-8) or raise the water level in other three branches (RS-5 to RS-7). For RS-5 to RS-7, as the water level of the south branch, the middle branch, and the north branch increased by the same range (from 15.5 m to 15.6 m), the discharge of the main branch increased to 336.51 m³/s, 344.21 m³/s, and 358.1 m³/s, respectively (

Table 7. Regulation effect comparison for difference water level scenarios.

RS-5	Waizhou	Main	North	Middle	South
Water level (m)	15.5	15.5	15.5	15.5	15.6
Demand discharge (m3/s)	488.7	341	5.1	60.4	61
Actual discharge (m3/s)	800	336.51	133.5	195.32	134.67
RS-6	Waizhou	Main	North	Middle	South
Water level (m)	15.5	15.5	15.5	15.6	15.5
Demand discharge (m3/s)	488.7	341	5.1	60.4	61
Actual discharge (m3/s)	800	344.21	136.57	127.62	191.6
RS-7	Waizhou	Main	North	Middle	South
Water level (m)	15.5	15.5	15.6	15.5	15.5
Demand discharge (m3/s)	488.7	341	5.1	60.4	61
Actual discharge (m3/s)	800	358.1	81.65	181.82	178.43
RS-8	Waizhou	Main	North	Middle	South
Water level (m)	15.5	15.4	15.5	15.5	15.5
Demand discharge (m3/s)	488.7	341	5.1	60.4	61
Actual discharge (m3/s)	800	384.55	97.65	157.98	159.82

). The evaluation of regulation scenarios (RS5-RS-8) is shown in

Table 8. Evaluation of regulation scenarios (RS5-RS-8).

Demand	RS-5	RS-6	RS-7	RS-8
Water level	✓	✓	✓	✓
Discharge	×	✓	✓	✓

, which indicates that RS6-RS8 could achieve the regulation objectives. However, the main branch’s discharge sensitivity to water level changes in the other three branches varies. Specifically, adjusting the water level of the north branch—closer to the main branch—has a more significant impact on the main branch’s discharge than adjusting that of the south branch, which is farther away. For RS-8, the discharge of the main branch was 384.55 m³/s, significantly higher than that of RS-5 to RS-7. To increase the main branch’s discharge, lowering its water level is more effective than raising the water level of the other three branches. This may be attributed to the fact that the discharge of each branch is primarily related to its water surface gradient. By lowering the water level upstream of the main branch’s gate, the water surface gradient of the main branch could be effectively increased. However, raising the water levels upstream of other branches’ gates has a very limited effect on increasing the water level in the upper reaches of the main branch, thus making it difficult to significantly enhance the discharge of the main branch.

4. Discussion

4.1. The Interaction Between the GJRT and Poyang Lake

Due to the significant annual water level fluctuation of Poyang Lake, the jacking and dropping effect of Poyang Lake on the GJRT is the main factor influencing the hydrodynamic processes in the GJRT. In this paper, we found that the water level of the GJRT decreased in October, while the upstream discharge decreased in September. It could be concluded that the water level in the GJRT decreased significantly under the influence of the dropping effect of Poyang Lake. Li et al. [22] conducted an analysis of the hydrodynamic processes in Poyang Lake and identified that the water level recession period primarily takes place in October. Furthermore, in recent years, the accelerated decline in Poyang Lake’s water levels has led to both the advancement and prolongation of the dry season—an occurrence that is likely to aggravate the drought-related challenges in the GJRT [23]. As Xingzi Station is a representative hydrological station in Poyang Lake, we analyzed the water level processes of Poyang Lake (Xingzi Station) from 1953 to 2023 (Figure 8). From 1953 to 2023, rising period of Poyang Lake was mainly from March to July, with no significant changes in either the rising rate or magnitude of the water level. Compared with the period before 2003, the water level of Poyang Lake during the recession period from 2003 to 2013 decreased; specifically, the recession rate increased after October. After 2013, the recession period of Poyang Lake advanced by 17 days compared with the period from 2003 to 2012, the recession rate increased significantly, and the maximum water level difference during the recession period could reach 2.5 m. Zhang et al. [24] also found that after the operation of Three Gorges Dam (2003), the low water period of Poyang Lake has been advanced and prolonged. The significant dropping effect caused by the rapid water recession of Poyang Lake on the GJRT fully demonstrates the necessity of constructing the GJRT water conservancy project group. Meanwhile, the regulation of the water conservancy project group serves as a crucial measure to mitigate the rapid water recession in the GJRT and cope with drought issues in this region.

4.2. Evolution of Diversion Characteristics of Multi-Branch Channels in the GJRTe

The multi-branch channels in the GJRT exhibit a distinct evolutionary trend: the main branch has been gradually developing with a continuous increase in its diversion ratio, while the other three branches (north branch, middle branch, and south branch) have shown a gradual shrinking trend [25]. Jia et al. [26] found that the progradation rate of middle reach of the GJRT decreased from 0.875 to 0.21 km/year from 1992 to 2015.

The change in diversion characteristics in the GJRT could be attributed to the thalweg elevation difference between each branch. As the reservoir construction upstream of the GJRT, the suspended sediment load in the Ganjiang River basin decreased continuously, which decreased by 48% in 1990–2002 and 75% in 2003–2014 [27]. In this case, the thalweg elevation of the GJRT decreased, with the main branch thalweg decreasing faster than other three branches [28]. As Waizhou Station is the inlet of the GJRT, we statistically evaluated the annual sediment load and runoff of Waizhou Station from 1952 to 2017 (Figure 9). From 1952 to 2018, in the case of the annual runoff remains basically unchanged, the annual sediment load began to decrease after 1985 and then showed a gradually declining trend. A further analysis of the annual sediment load variations reveals that the changes in the annual sediment load of the GJRT can be divided into three phases as follows: from 1952 to 1984, the annual sediment load was approximately 1200 × 10⁴ t; from 1985 to 1998, the annual sediment load decreased to 700 × 10⁴ t; from 1998 to 2017, the annual sediment load further decreased to 280 × 10⁴ t.

4.3. Impacts of Rapid Water Recession on the Ecology and Economy in the GJRT

The rapid water recession in the GJRT would significantly impact the irrigation, navigation, and ecology in the GJRT. The elevations of the bottom plate of the main irrigation intake around Poyang Lake range from 12 to 14 m, and irrigation area covered by these intakes exceeds 66,667 hectares. Figure 10 shows the water recession process from 1 September to 31 October. It could be found that without regulation, the water levels of all branches in the GJRT were lower than the elevation of irrigation intake by the end of October, resulting in widespread difficulties in agricultural water abstraction.

As the main waterway of both the GJRT and Poyang Lake, the main branch has a planned navigable water depth of 4.5 m. Figure 11 presents the water depths of the cross-section of the main branch at the end of October under natural conditions and after regulation. Under natural conditions, the navigable width of the main branch is less than 200 m after the water recession. However, with the regulation of the main branch water conservancy project, the navigable width can reach 400 m. This indicates that setting the regulated water level in the GJRT at 15.5 m is of great significance for ensuring the navigation of Poyang Lake.

The north branch and middle branch of the GJRT are connected to the wetland of Poyang Lake, with many sub-lakes (Figure 12). Xu [29] found that after 2003, the average monthly water levels declined by 0.84 m, which disrupted the growth cycles of wetland vegetation. Under the condition of rapid water recession in the GJRT, the north branch and middle branch were almost dried up by the end of October (Figure 7). As important water sources for Poyang Lake’s sub-lakes, the drying-up of these two branches will exacerbate the drought stress of the sub-lakes.

4.4. Limitations

Optimization algorithms serve as effective tools for formulating regulating scenarios in water conservancy projects. In the future, by integrating optimization algorithms with numerical regulating experiments, superior regulatory scenarios for water conservancy projects in the GJRT can be developed.

5. Conclusions

In response to the dry event frequently occurring during the dry season in the multi-branch channels at the tail of the Ganjiang River, this paper proposes a joint regulation strategy of the water conservation project group. This regulation strategy considers the hydrological situation in dry season and diversion characteristics of multi-branch channels, achieving the maintenance of both water level and discharge for each branch during the dry seasons. The following are the main conclusions:

(1)

Affected by the accelerated water recession of Poyang Lake in recent years, the Ganjiang River tail experiences severe drought during the dry period, driven by the dropping effect of Poyang Lake. Additionally, the uneven incision of the multi-branch channels in the Ganjiang River tail has led to irrational flow distribution among the four branches, which further exacerbates the drought conditions in this region.

(2)

Affected by the jacking effect of Poyang Lake, the water level decline in the Ganjiang River tail lags the flow decline. Activating the project group for water storage at the initial stage of water level decline can effectively utilize the water resources at the end of the flood season and quickly reach the target water level of the Ganjiang River tail.

(3)

Compared with equal water level scenarios, different water level scenarios could effectively improve the diversion ratio of the multi-branch channels. When conducting joint regulation of water conservancy project group in the multi-branch channels of the Ganjiang River tail, discharge of one branch responses differently to water levels of different branches. Only by analyzing the response relationship between discharge and water level while considering the interaction between different branches can we improve the regulation efficiency.