https://orcid.org/0009-0008-0183-8549
ABSTRACT
Water infrastructure affects the quantity, quality, and environment of water. During reservoir construction, only the social-economic benefits were considered and the health of the downstream river was neglected. Due to the severe downstream river system, the operators and managers faced the challenge of the downstream ecosystem. Many scholars are concerned with the reservoir's optimum functioning, which considers social and economic advantages. Numerous significant opinions and research findings are being presented on ecological scheduling at the moment, both in China and overseas, which surely deepen the connotations of ecological scheduling and serve as the research basis and critical reference of this work. The influence of hydropower development on hydrological conditions, the categorization and calculation of environmental runoff, and the building and solution of ecological scheduling models are all discussed and researched in this work. The study examined how the ChiTan hydropower project influenced the Jinxi River's flow. The total hydrologic alteration calculated using RVA is 50.53%. The component analysis is also utilized to eliminate the statistical redundancy in the hydrologic indicators. The indicators’ monthly flow for July and August's mean minimum flow for 7 days are relevant indicators for the Jinxi River basin's hydrologic modification regime.
HIGHLIGHTS
This paper discusses how the ChiTan hydropower project influenced the Jinxi River's flow.
The influence of hydropower development on hydrological conditions, the categorization.
The 33 characteristics utilized in IHA are interconnected and may be reduced to representative river indicators using PCA. This work used PCA to determine the smallest number of usual indicators of hydrologic change in the Jinxi River Basin.
INTRODUCTION
Many factors, such as politics, economy, and ecology, have a role in the overall growth of cascade hydroelectric energy in the basin to address the population's needs. Demand for hydropower, flood control, shipping, drinking water, irrigation, and electricity is expected to rise. Water conservation projects will continue to be built to improve the natural environment and, hence, our people's living conditions. One of the most used production methods, artificial flood control and power generation methods, has been installed in China's major rivers. Even though China has created a mechanism for assessing the environmental effect of water conservation project development, the state also passed the Environmental Assessment Law in 2002. Hatamkhani et al. (2023a) examines the incorporation of ecosystem services into optimization models and decision-making processes for hydropower projects to promote sustainable development. The method measures the environmental impact and incorporates the benefits of hydropower and the negative consequences on ecosystem services into the overall goal. Hatamkhani et al. (2022) examines the economic worth of wetlands, which are diverse and productive ecosystems, and their relationship with agricultural development. The paper proposes that including the financial cost of wetland ecosystem services in managing agricultural water usage can result in efficient water distribution, meet environmental needs, and facilitate sustainable basin planning. Spanoudaki et al. (2022) introduces a methodological instrument for assessing the hydropower capacity of a reservoir by utilizing hydrological data. The use of Bayesian analysis, stochastic analysis, Monte Carlo ensemble technique, synthetic river flow time series, and cladogram metre achieves the simulation of reservoir water balance and estimation of hydropower potential.
On the other hand, large-scale water conservation programmes negatively affect the biological environment at various temporal and spatial scales at the ecosystem level. Noise is a subject that has received minimal investigation. There is insufficient scientific evidence to support the development, operation, and management of water conservation projects and biodiversity. Conservation and vital biological resources are coordinated at a macro level. It also puts in place a robust multidimensional regulatory decision-making framework. Outside of the basin, at the crossroads of ecological hydrology response mechanisms and other disciplines, where changes in environmental variables have a negative influence on the basin's ecological environment, the water conservancy department also lacks efficient control mechanisms to enhance the environment and prevent or eliminate these negative consequences because research in this sector is far behind. The integration of environmental impacts into optimizing and making decisions about hydropower projects was studied by Hatamkhani et al. (2023b). To this end, a comprehensive simulation–optimization model is created, including positive and negative ecological effects on economic goals. Hatamkhani et al. (2021) developed a simulation–optimization model for the simultaneous design and operation of hydropower dams using WEAP and invasive weeds optimization. The rivalry between society's water requirements and the need to maintain ecosystems has intensified due to diminishing water supplies and increasing demand. Waterbirds and vegetation can be used as bioindicators to assess the ideal environmental flow, which guarantees the ecological state and functionality of the wetland (Hatamkhani & Moridi 2023). Yousefi & Moridi (2022) employed SWAT and NSDE to create a simulation–optimization model to devise effective ways to minimize climate change impacts. The model optimized agricultural revenue while minimizing nutrient discharge into the Minab reservoir. Kuriqi et al. (2021) examined the ecological consequences of three primary minor run-of-river hydropower designs: dam-toe, diversion weir, and bondage schemes. The study concluded that diversion weir and bondage hydropower schemes have a lower level of environmental friendliness, whereas dam-toe hydropower schemes exhibit a higher level of environmental friendliness. With the strengthening of the national economy and the western growth driven by water conservation, the detrimental effects generated by building water conservancy projects will also be short if there is no adequate countermeasure. It may become a bottleneck in the future, preventing the implementation of national sustainable development plans. As a result, the watershed is used as a unit to investigate the ecological functions of the watershed and their relationships at various temporal and geographical scales. Yes, management is essential for increasing and using the basin's water resources and attaining man and nature's synchronized growth. A group of ecologists introduced the ‘natural flow regime’ concept in the late 1990s, which is more advantageous to the aquatic ecosystem (Poff et al. 1997). Seasonal water, sediment, and nutrient flux variations may be seen in rivers and streams. Hydrological regimes are directly affected by human activities such as energy generation, flood prevention, agriculture, and inland water transportation. Reservoirs are built to produce hydroelectric power, provide water, and give recreational opportunities (Guo et al. 2021). In reservoir operation, streamflow forecasting is critical. The most essential problem in water resource management is the proper functioning of reservoir systems. A significant difficulty for water resource management in the twenty-first century is that environmental and human demands must be addressed in formulating reservoir and stream management plans using regime-based approaches (Chang et al. 2014). Water conservation projects often serve several purposes, including flood control, electricity production, ecological demands, and water supply. Consequently, a single-objective operation focused entirely on water supply or power production can no longer meet current needs. Therefore, it's become evident that numerous objectives must be met simultaneously to maximize the total benefits of water conservation programmes (Sun et al. 2018).
The Jinxi River Basin water resources system is an enormous, open, and complicated system whose optimal design and management depend on a specific natural, economic, and social environment, particularly in China. The basin's natural ecosystem is changing faster with the addition of human activity. Water conservation schemes obstruct and constrain large rivers. Flow is a physical concept. The transformation of natural runoff energy and the transfer of resources are at the heart of the cascade water resources system's progress.
Background and research objectives
In China, there is a significant time and spatial imbalance in the distribution of water resources. Regarding temporal distribution, China's water resources have an uneven distribution throughout the year and considerable interannual fluctuations, resulting in significant variances in most Chinese rivers' wet and dry seasons. First, China's adverse congenital characteristics of water resources have hampered the development of water resources; second, China is prone to recurrent floods and droughts. Reservoir building and operation can reduce flood and drought calamities and address supply and demand contradictions created by unequal temporal and geographical distribution of water resources. China have developed 98,112 reservoirs of different sizes as of 2020. With China's continued expansion of hydroelectric energy, the question of how to rationally optimize joint reservoir operation and pay for advantages has become more critical in the business. There are still specific issues in the deterministic optimum functioning of cascade reservoirs that need to be addressed:
(i) As the time step is reduced and the calculation period is increased, the link between upstream and downstream hydraulic and electric power is strengthened, and the difficulty of constraint treatment increases, resulting in an exponential rise in the problem's complexity.
(ii) As society evolves, the operation of cascade reservoirs must consider multiple objectives regarding social economy and environmental Envi. The situation becomes much more complicated as the number of goals grows.
This work sparked a flurry of research into hydrological change, streamflow forecasting, reservoir ecological scheduling, and reservoir management with ecological flow in mind. It calculates the hydrologic alteration of the Jinxi River basin due to the Chi Tan hydropower station, identifying representative indicators of the Jinxi River basin using principle component analysis (PCA), and calculating the minimum and suitable flow of the Jinxi River to incorporate it into the optimization model using the environmental flow calculation method. Minjiang River is the upper reaches of Jinxi River and is situated between 26° 18′ 28.8″ ∼ 27° 7′ 40.8″ N and 116° 26′ 52.8″ ∼ 117° 5′ 24″ E. The Minjiang River, or the Mother River of Fujian, is the biggest in Fujian Province, with a length of 577 km. It is situated in China's southeast coastline region. The river rises in the Wuyi Mountains, runs through 36 counties and districts, and then past the densely populated suburbs and inner city of Fuzhou, Fujian Province's capital, before emptying into the East China Sea. The Minjiang River is a major supplier of drinking and recreational water and agricultural irrigation water. The whole river basin drains around 61,000 km2, almost half of Fujian Province's total area. Around 16 million people live in the region, with 4 million living in metropolitan areas. The NSGA-II algorithm determines the multiobjective nature of reservoir operation, considering ecological demand and benefits. The Chi Tan hydropower plant is investigated based on the theory stated in the paper.
In the 1960s, foreign academics started researching the best way to operate cascade reservoirs. In 1968, Larson (1968) used incremental dynamic programming to improve reservoir operations, avoiding the poor efficiency of dynamic programming in multidimensional optimization problems. In 1970, Bellman (Bellman & Zadeh 1970) developed the fuzzy emotional programming technique (FDP), which is suited for dealing with fuzzy goal or constraint boundary conditions and provides a unique concept for reservoir group optimal operation. Arvanitidits & Rosing (1970) presented composite representation (CR) to tackle the reservoir group system optimization issue in the same year. In 1974, Askew (1974) introduced a dynamic programming approach that used safety restrictions to address the problem that the reservoir operation outcome of stochastic dynamic programming may cause system failure. Based on Lagrange duality theory Rossman (1977) introduced a stochastic dynamic programming model bound by system performance reliability to establish reservoir operating rules that may optimize projected net revenue in 1977.
In 1978, the single-aim reservoir operation model was effectively used for stochastic dynamic programming (Viramontes & Hamilton 1978). To address the problem of multiobjective reservoir operation, Tauxe et al. (1979) developed a multiobjective dynamic programming approach. Turgeon (1980) proposed in 1980 to decompose the reservoir group optimal operation model's multidimensional state variables into multiple stochastic optimization subproblems with two state variables, which are also solved by dynamic programming and are primarily used to formulate the reservoir group's weekly strategy. Maidment & Chow (1981) presented the random state variable dynamic programming approach for developing the Watasheamu dam operating plan in 1981, combining the state notion, dynamic programming, and Markov chain analysis utilized in state space modelling. Yeh (1985) summarized and implemented several linear programming approaches to reservoir optimal operation in 1985. The cascade dynamic programming method, a backwards-moving stagewise optimization process, was introduced by Foufoula-Georgiou & Kitanidis (1988). It successfully eliminates the ‘disaster of dimensionality’ issue using a crude discrete network and the Hermite interpolation algorithm to increase operation efficiency.
Saad & Turgeon (1988) recommended using PCA to reduce the dimension of a high-dimensional reservoir group's optimum operation problem and optimize it. They applied it to the optimal operation of the La Grande River's cascade reservoirs. Consoli et al. (2008) suggested applying the nonlinear programming constraint technique to address reservoir operation optimization for irrigation. Nagesh Kumar et al. (2010) and colleagues used folded dynamic programming (FDP) to develop the best reservoir operating plan 2010 (Kimmany et al. 2020). To maximize the hydropower production of cascade reservoir operations in the Nam Ngum River Basin in 2020, a mixed integer nonlinear programming (MINLP) optimization model was developed. In his paper, Professor Xiang Feng Huang discusses the chaotic optimal operation of hydropower, considering ecological needs. With high efficiency that can produce more power, the price of electricity can be suitable for the present market when considering downstream water flow as an ecological need (Huang et al. 2010). Because the distribution and variety of the Pareto-optimal sets of the solution are acceptable, the NSGA-II technique is demonstrated to be appropriate for solving the reservoir-scheduling model.
METHODOLOGY
The lowest ecological discharge is the average flow of the driest 7 days with a 90% certainty rate in the river channel. The criteria of the 7q10 technique are typically high, but the water resources condition and degrees of social and economic development in various parts of China vary significantly. As a result, the lowest monthly average flow in the previous 10 years or the lowest monthly average flow beneath the 90% guarantee rate may be utilized for the minimum ecological flow of general rivers. This study used daily water flow data for ChiTan hydrological stations in the Jinxi River basin, which provided the observed data.
Hydrological methods
The hydrological technique is based on streamflow data from the past. Even if the results aren't exact, they can be obtained quickly. This strategy is widely acknowledged as being helpful during the planning stage of water-related projects. The most common process is the Tennant (or modified Tennant) approach. The second most often used approaches include various flow duration surpasses percentiles (e.g., Q95, Q75), single low flow indices (e.g., 7Q10, 7Q2), and the minimum continuous 30-day mean discharge approach.
The mathematical formulation of hydrologic indicators was explored by Barbali et al. (‘Trends of indicators of hydrological alterations’ 2014).
is the indicator of group 1, for month m (m3/s); n is the ordinal number of the month, 1 ≤ n ≥ 12; m is the number of days in a month n; 
is the daily average discharge (m3/s), ith day of nth month.
Example of monthly average flow.
Indicator Group 2: Annual severe water conditions: magnitude and duration
is the indicator from group 2 for minima (m3/s), 
; m is the duration (days), 
; 
is the daily average discharge (m3/s), jth day of the year.Indicator Group 3: Timing of annual extreme water conditions
, 
are the indicators of hydrologic alteration from group 3; I is the ordinal number of a day within a year 
; 
is the daily average discharge (m3/s), jth day of the year; 
is the maximal recorded daily average discharge (m3/s) during the year; 
is the minimal recorded daily average discharge (m3/s) during the year.Indicator Group 4: Frequency and duration of high/low pulses
, 
are the number of high and low pulses; 
, 
are the mean duration of high and low pulses (days); I is the ordinal number of the day within a year 
; 
is the daily average discharge (m3/s), jth day of the year; 
is the discharge of 25% duration (m3/s); 
is the discharge of 75% duration (m3/s).
Indicator Group 5: Rate/frequency of water condition changes
, 
and 
are the indicators of group 5 (
); j is the ordinal number of the day within a year 
; 
is the daily average discharge (
); jth day of the year.Principle component analysis
PCA is a multivariate statistical method that uses an orthogonal transformation to turn a set of correlated variables into a set of orthogonal, uncorrelated axes called principal components (PCs). The 33 factors utilized in indicators of hydrologic alteration (IHA) are interconnected and may be simplified to river-specific indicators. ‘Even though the original variables are in, the same units, the variances may range substantially, typically because they are tied to their names,’ Jackson continues (2005). The correlation matrix should be used if this provides excessive weight to particular variables.’ The same procedure is used for IHA parameters, which are stated in several units: frequency, duration, timing, amplitude, and rate of occurrences.
The correlation matrix for PCA is created using data from 33 IHA parameters examined by the programme (no zero-day flow). The eigenvalue is derived from a 33-by-33 correlation matrix for Pearson and Spearman techniques. The selection of PCs is done based on eigenvalues. Kaiser (1974) stated the preservation of meaningful elements is done based on Kaiser–Guttman Criteria, with PCs with eigenvalues higher than 1.0 (ʎ > 1.0) (Kaiser 1974). The selection of subspaces from retained main components is done via orthogonal rotation. The original subspace's overall maximum Variance stays constant after the cycle, but it is spread more equally across the rotated segments than before (DeSarbo et al. 2007). Kaiser's orthogonal Varimax rotation is used to get factor loadings (Kaiser 1958). The PC is represented by a single variable with the most significant absolute loadings on each chosen PC. The Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy (MSA) is still more effective since the number of IHA indices kept is still higher (MSA). Table 1 is performed to reduce the number of indices, and only marvellous class indices are selected.
Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy (MSA)
| Class | Marvellous | Meritorious | Middling | Mediocre | Miserable | Unacceptable |
|---|---|---|---|---|---|---|
| Loading range | 0.9–1.0 | 0.8–0.89 | 0.7–0.79 | 0.6–0.69 | 0.5–0.59 | 0.0–0.49 |
| Class | Marvellous | Meritorious | Middling | Mediocre | Miserable | Unacceptable |
|---|---|---|---|---|---|---|
| Loading range | 0.9–1.0 | 0.8–0.89 | 0.7–0.79 | 0.6–0.69 | 0.5–0.59 | 0.0–0.49 |
The role of PCA is to select the minimum number of representative IHA of the Jinxi River basin. The principal component analysis using Pearson and Spearman methods was performed using the XLSTAT software.
Ecological flow
The importance of instream biological flow in determining river health and reservoir operation cannot be overstated. The computation of ecological flow is the foundation of reservoir ecological scheduling. The minimal environmental flow calculation and the appropriate ecological flow calculation are discussed in this section.
Minimum ecological flow
The concept of minimum ecological flow states that a particular amount of minimum flow must be present in a flowing freshwater ecosystem such as a river or an estuary to sustain the flowing freshwater ecosystem's ecological quality.
The minimal average flow of 30 consecutive days in the river channel within a specified assurance rate is the minimum ecological runoff at various times, depending on upstream reservoir control, inflow circumstances, seasons, etc.
The goal of the 7q10 approach is to reduce polluting sources' emissions. The lowest ecological discharge is the average flow of the driest 7 days with a 90% certainty rate in the river channel. The criteria of the 7q10 technique are typically high, but the water resources condition and degrees of social and economic development in various parts of China vary significantly. As a result, the lowest monthly average flow in the previous 10 years or the lowest monthly average flow beneath the 90% guarantee rate may be utilized for the minimum ecological flow of general rivers (Caissie & El-Jabi 1995).
This technique assumes that, in the natural world, if the actual minimum monthly average runoff has not caused substantial and lasting harm to the ecosystem in the past, this flow may be considered the minimal ecological runoff, and aquatic creatures can adapt to and spend time safely in it.
Suitable ecological flow
is the deviation from the mean, 
. A new quantity can be defined as: 
, where s is the standard deviation of the data, and the former equation can be rewritten as:
. With a return period of T, Equation (13) can be written as:
is the coefficient of variation, and 
is the frequency factor depending on the probability distribution and return period T. Equation (14) is the working equation for frequency analysis, which can be used to calculate the design value 
for given design level T and probability distribution.The concept of suitable ecological flow is of paramount importance, as it goes beyond the minimum ecological flow needed to ensure the thriving and prosperity of the riverine ecosystem. It takes into account the varying flow requirements during different seasons to mimic natural hydrological patterns. These requirements, specific to various species, encompass activities such as mating, feeding, and migration, and the preservation of river services, such as sediment transportation, nutrient cycling, and floodplain connection. The calculation of appropriate ecological flow is a comprehensive process that leaves no stone unturned. It involves identifying the criteria that establish a robust ecosystem, such as the well-being of fish populations and the diversity of plant species. Appropriate flow criteria are then determined based on these indicators, ensuring the necessary flows for fish spawning, determining the proper wetland inundation levels for bird breeding, and establishing seasonal flows to preserve plant diversity. Hydrological data, such as past flow records, is gathered to comprehend the natural fluctuations in water flow. Ecological models are utilized to replicate various flow patterns, and the extent to which each scenario aligns with the ecological objectives is analysed. A comparison with the real flows is then conducted, and any discrepancies are detected by evaluating present and anticipated water movements compared to the simulated optimal ecological water movements.
Reservoir optimal dispatching model
The model is based on theories related to multiobjective hydropower dispatching and the reality of the ChiTan hydropower plant. The multiobjective reservoir ecological operation model's goal is as follows:
Maximization of power generation
is the total annual generation of hydropower stations, 108 kWh, 
is the power generation coefficient (
), 
is the turbine release water discharge for power generation at period t, m3/s, 
is the average head at period t, m. 
is the duration of the period, s.Minimization of water shortage in the river
, 
is the water diversion and power generation flow of the reservoir, 
is the wastewater from the dam, 
is the minimum ecological flow of the river at the time period t, 
is the suitable ecological flow of the river at the time period t, 
is the discharge from the ecological sluice install in the dam, and 
is the average annual discharge of the river at the dam site; all are measured in 
.Assessment of hydrologic alteration caused by Chi Tan station in Jinxi River Basin
By considering the inherent hydrologic variability of rivers, Richter et al. (1997) and Arvanitidits & Rosing (1970) presented the range of variability techniques for setting river ecosystem management objectives based on streamflow (Richter et al. 1997). This technique characterizes streamflow records using 33 distinct hydrological characteristics to analyse hydrologic changes: Size, timing, frequency, duration, and rate of change of streamflow (Table 3). Richter et al. (1998) showed how to apply the RVA to measure hydrologic change throughout a river basin using accessible stream gauge sites. The RVA approach is a watershed moment in hydrologic change assessment (Richter et al. 1998). It has been used in several studies to assess hydrologic changes caused by human intervention, such as dam building.
RVA analysis.
In contrast, a negative value indicates a drop in the frequency of values. Positive hydrologic alteration values were found in seven parameters, whereas negative hydrologic alteration values were found in 20, and no alteration or zero alteration was found in 5. The range categorization for the change values is shown in Table 2.
List of hydrological parameters with alteration rate
| IHA group (I) | RVA (%) | IHA group (II) | RVA (%) | IHA group (III) | RVA (%) |
|---|---|---|---|---|---|
| Apr | 23.53 | 1-day minimum | −47.06 | Date of minimum | −29.41 |
| May | −29.41 | 3-day minimum | −100 | Date of maximum | 23.53 |
| Jun | −29.41 | 7-day minimum | −82.35 | ||
| Jul | −47.06 | 30-day minimum | 23.53 | IHA group (IV) | |
| Aug | −47.06 | 90-day minimum | 5.882 | Low pulse count | −64.71 |
| Sep | 5.882 | 1-day maximum | 23.53 | Low pulse duration | −26.47 |
| Oct | −11.76 | 3-day maximum | 5.882 | High pulse count | −55.88 |
| Nov | −11.76 | 7-day maximum | 5.882 | High pulse duration | 32.35 |
| Dec | −11.76 | 30-day maximum | −29.41 | ||
| Jan | −11.76 | 90-day maximum | 47.06 | IHA group (V) | |
| Feb | 5.882 | Base flow index | 5.882 | Rise rate | −29.41 |
| Mar | 23.53 | Fall rate | 23.53 | ||
| Number of reversals | −26.47 |
| IHA group (I) | RVA (%) | IHA group (II) | RVA (%) | IHA group (III) | RVA (%) |
|---|---|---|---|---|---|
| Apr | 23.53 | 1-day minimum | −47.06 | Date of minimum | −29.41 |
| May | −29.41 | 3-day minimum | −100 | Date of maximum | 23.53 |
| Jun | −29.41 | 7-day minimum | −82.35 | ||
| Jul | −47.06 | 30-day minimum | 23.53 | IHA group (IV) | |
| Aug | −47.06 | 90-day minimum | 5.882 | Low pulse count | −64.71 |
| Sep | 5.882 | 1-day maximum | 23.53 | Low pulse duration | −26.47 |
| Oct | −11.76 | 3-day maximum | 5.882 | High pulse count | −55.88 |
| Nov | −11.76 | 7-day maximum | 5.882 | High pulse duration | 32.35 |
| Dec | −11.76 | 30-day maximum | −29.41 | ||
| Jan | −11.76 | 90-day maximum | 47.06 | IHA group (V) | |
| Feb | 5.882 | Base flow index | 5.882 | Rise rate | −29.41 |
| Mar | 23.53 | Fall rate | 23.53 | ||
| Number of reversals | −26.47 |
As indicated in Table 3, the IHA indices of group I were adjusted from 5.88 to 47.06%. According to the table, there is a lot of alteration in the mean monthly value in July and August; moderate alteration in the mean monthly value in March, April, May, and June; less alteration in the mean monthly value in January, October, November, and December; and tiny alteration in the mean monthly value in February and September.
Basic parameters of each power station
| Indicator name | Chi Tan | Liang Qian | Dayan | Huangtan | Kongtou | Fancy | Gaotang | Mowu | Guiling |
|---|---|---|---|---|---|---|---|---|---|
| Regulation performance | Incomplete annual adjustment | Nothing | Nothing | Nothing | Nothing | Nothing | Nothing | Nothing | Nothing |
| Installed capacity/MW | 200 | 30 | 32 | 30 | 40.5 | 39.6 | 42 | 33 | 7.6 |
| Guaranteed output/MW | 36.4 | 8.69 | 8.4 | 8.9 | 10.1 | 9.9 | 9.91 | 8.4 | 6.9 |
| Normal water level/M | 275 | 211.2 | 197.8 | 193.8 | 175.8 | 161 | 146 | 133 | 130 |
| Dead water level/M | 245 | 209.2 | 196.3 | 191.3 | 174 | 160 | 144.8 | 128.8 | 123.8 |
| Regulating storage capacity/10,000 m3 | 66,500 | 293 | 267 | 612 | 699 | 400 | 516 | 475 | 110 |
| Minimum ecological flow/(M3 · s-1) | 23.6 | 23.77 | 25.05 | 25.82 | 26.68 | 28.81 | 33.09 | 34.43 | 35.66 |
| Maximum power generation flow/(M3 · s-1) | 400 | 326.1 | 378 | 369.3 | 369.3 | 420 | 512.2 | 488 | 488 |
| The ratio of the electricity price of each power station to that of the leading power station | 1.00 | 1.22 | 1.47 | 1.47 | 1.59 | 1.22 | 1.63 | 1.63 | 1.22 |
| Indicator name | Chi Tan | Liang Qian | Dayan | Huangtan | Kongtou | Fancy | Gaotang | Mowu | Guiling |
|---|---|---|---|---|---|---|---|---|---|
| Regulation performance | Incomplete annual adjustment | Nothing | Nothing | Nothing | Nothing | Nothing | Nothing | Nothing | Nothing |
| Installed capacity/MW | 200 | 30 | 32 | 30 | 40.5 | 39.6 | 42 | 33 | 7.6 |
| Guaranteed output/MW | 36.4 | 8.69 | 8.4 | 8.9 | 10.1 | 9.9 | 9.91 | 8.4 | 6.9 |
| Normal water level/M | 275 | 211.2 | 197.8 | 193.8 | 175.8 | 161 | 146 | 133 | 130 |
| Dead water level/M | 245 | 209.2 | 196.3 | 191.3 | 174 | 160 | 144.8 | 128.8 | 123.8 |
| Regulating storage capacity/10,000 m3 | 66,500 | 293 | 267 | 612 | 699 | 400 | 516 | 475 | 110 |
| Minimum ecological flow/(M3 · s-1) | 23.6 | 23.77 | 25.05 | 25.82 | 26.68 | 28.81 | 33.09 | 34.43 | 35.66 |
| Maximum power generation flow/(M3 · s-1) | 400 | 326.1 | 378 | 369.3 | 369.3 | 420 | 512.2 | 488 | 488 |
| The ratio of the electricity price of each power station to that of the leading power station | 1.00 | 1.22 | 1.47 | 1.47 | 1.59 | 1.22 | 1.63 | 1.63 | 1.22 |
Factor loading for Spearman method on retained seven components
| Parameter | F1 | F2 | F3 | F4 | F5 | F6 | F7 |
|---|---|---|---|---|---|---|---|
| Jan | 0.059 | −0.045 | 0.315 | 0.088 | −0.176 | 0.852 | 0.055 |
| Feb | 0.172 | −0.038 | 0.198 | 0.344 | −0.166 | 0.837 | 0.102 |
| Mar | 0.014 | −0.168 | 0.058 | 0.725 | 0.010 | 0.506 | −0.015 |
| Apr | 0.182 | 0.107 | 0.283 | 0.731 | −0.302 | −0.023 | 0.173 |
| May | 0.493 | −0.146 | 0.174 | 0.620 | 0.046 | −0.019 | −0.166 |
| Jun | 0.681 | 0.224 | 0.301 | 0.428 | −0.069 | 0.175 | 0.096 |
| Jul | 0.626 | 0.624 | 0.064 | 0.146 | 0.134 | 0.072 | 0.034 |
| Aug | 0.227 | 0.877 | 0.120 | 0.019 | 0.071 | 0.002 | 0.098 |
| Sep | 0.125 | 0.860 | 0.026 | 0.191 | 0.116 | −0.048 | 0.165 |
| Oct | 0.143 | 0.799 | 0.055 | 0.002 | 0.362 | −0.105 | 0.004 |
| Nov | 0.031 | 0.408 | 0.258 | −0.108 | 0.794 | −0.064 | −0.016 |
| Dec | 0.172 | 0.515 | 0.297 | 0.018 | 0.717 | 0.021 | 0.023 |
| 1-day min | 0.350 | 0.110 | 0.776 | 0.182 | −0.022 | 0.120 | −0.047 |
| 3-day min | 0.194 | 0.012 | 0.945 | 0.005 | −0.065 | 0.103 | −0.008 |
| 7-day min | 0.131 | 0.006 | 0.941 | 0.061 | 0.120 | 0.094 | 0.112 |
| 30-day min | 0.079 | 0.255 | 0.801 | 0.033 | 0.309 | 0.175 | 0.236 |
| 90-day min | 0.305 | 0.448 | 0.390 | 0.055 | 0.693 | −0.086 | 0.071 |
| 1-day max | 0.897 | 0.134 | 0.087 | −0.064 | 0.129 | −0.083 | 0.094 |
| 3-day max | 0.959 | 0.097 | 0.119 | −0.049 | 0.088 | −0.002 | 0.011 |
| 7-day max | 0.951 | 0.077 | 0.115 | 0.070 | 0.151 | 0.106 | −0.009 |
| 30-day max | 0.886 | 0.185 | 0.202 | 0.290 | 0.098 | 0.083 | 0.022 |
| 90-day max | 0.795 | 0.249 | 0.245 | 0.452 | −0.002 | 0.058 | 0.015 |
| Base flow | −0.570 | −0.405 | 0.422 | −0.498 | −0.043 | −0.132 | 0.031 |
| Date min | −0.196 | −0.062 | 0.298 | 0.009 | −0.791 | 0.205 | −0.133 |
| Date max | 0.062 | 0.278 | 0.144 | −0.122 | 0.150 | 0.044 | 0.868 |
| Lo pulse | −0.334 | −0.584 | −0.097 | −0.139 | −0.127 | −0.293 | 0.137 |
| Lo pulse L | −0.080 | 0.393 | −0.191 | −0.149 | 0.086 | 0.566 | −0.400 |
| Hi pulse | −0.063 | 0.594 | −0.125 | −0.318 | 0.416 | −0.098 | 0.145 |
| Hi pulse L | 0.073 | 0.105 | −0.203 | 0.771 | −0.068 | 0.031 | −0.239 |
| Rise rate | 0.157 | 0.568 | 0.184 | 0.593 | 0.246 | 0.190 | 0.256 |
| Fall rate | −0.146 | −0.487 | −0.160 | −0.593 | −0.310 | −0.248 | −0.094 |
| Reversals | 0.059 | −0.045 | 0.315 | 0.088 | −0.176 | 0.852 | 0.055 |
| Parameter | F1 | F2 | F3 | F4 | F5 | F6 | F7 |
|---|---|---|---|---|---|---|---|
| Jan | 0.059 | −0.045 | 0.315 | 0.088 | −0.176 | 0.852 | 0.055 |
| Feb | 0.172 | −0.038 | 0.198 | 0.344 | −0.166 | 0.837 | 0.102 |
| Mar | 0.014 | −0.168 | 0.058 | 0.725 | 0.010 | 0.506 | −0.015 |
| Apr | 0.182 | 0.107 | 0.283 | 0.731 | −0.302 | −0.023 | 0.173 |
| May | 0.493 | −0.146 | 0.174 | 0.620 | 0.046 | −0.019 | −0.166 |
| Jun | 0.681 | 0.224 | 0.301 | 0.428 | −0.069 | 0.175 | 0.096 |
| Jul | 0.626 | 0.624 | 0.064 | 0.146 | 0.134 | 0.072 | 0.034 |
| Aug | 0.227 | 0.877 | 0.120 | 0.019 | 0.071 | 0.002 | 0.098 |
| Sep | 0.125 | 0.860 | 0.026 | 0.191 | 0.116 | −0.048 | 0.165 |
| Oct | 0.143 | 0.799 | 0.055 | 0.002 | 0.362 | −0.105 | 0.004 |
| Nov | 0.031 | 0.408 | 0.258 | −0.108 | 0.794 | −0.064 | −0.016 |
| Dec | 0.172 | 0.515 | 0.297 | 0.018 | 0.717 | 0.021 | 0.023 |
| 1-day min | 0.350 | 0.110 | 0.776 | 0.182 | −0.022 | 0.120 | −0.047 |
| 3-day min | 0.194 | 0.012 | 0.945 | 0.005 | −0.065 | 0.103 | −0.008 |
| 7-day min | 0.131 | 0.006 | 0.941 | 0.061 | 0.120 | 0.094 | 0.112 |
| 30-day min | 0.079 | 0.255 | 0.801 | 0.033 | 0.309 | 0.175 | 0.236 |
| 90-day min | 0.305 | 0.448 | 0.390 | 0.055 | 0.693 | −0.086 | 0.071 |
| 1-day max | 0.897 | 0.134 | 0.087 | −0.064 | 0.129 | −0.083 | 0.094 |
| 3-day max | 0.959 | 0.097 | 0.119 | −0.049 | 0.088 | −0.002 | 0.011 |
| 7-day max | 0.951 | 0.077 | 0.115 | 0.070 | 0.151 | 0.106 | −0.009 |
| 30-day max | 0.886 | 0.185 | 0.202 | 0.290 | 0.098 | 0.083 | 0.022 |
| 90-day max | 0.795 | 0.249 | 0.245 | 0.452 | −0.002 | 0.058 | 0.015 |
| Base flow | −0.570 | −0.405 | 0.422 | −0.498 | −0.043 | −0.132 | 0.031 |
| Date min | −0.196 | −0.062 | 0.298 | 0.009 | −0.791 | 0.205 | −0.133 |
| Date max | 0.062 | 0.278 | 0.144 | −0.122 | 0.150 | 0.044 | 0.868 |
| Lo pulse | −0.334 | −0.584 | −0.097 | −0.139 | −0.127 | −0.293 | 0.137 |
| Lo pulse L | −0.080 | 0.393 | −0.191 | −0.149 | 0.086 | 0.566 | −0.400 |
| Hi pulse | −0.063 | 0.594 | −0.125 | −0.318 | 0.416 | −0.098 | 0.145 |
| Hi pulse L | 0.073 | 0.105 | −0.203 | 0.771 | −0.068 | 0.031 | −0.239 |
| Rise rate | 0.157 | 0.568 | 0.184 | 0.593 | 0.246 | 0.190 | 0.256 |
| Fall rate | −0.146 | −0.487 | −0.160 | −0.593 | −0.310 | −0.248 | −0.094 |
| Reversals | 0.059 | −0.045 | 0.315 | 0.088 | −0.176 | 0.852 | 0.055 |
Extraction method: principal component analysis.
Rotation method: varimax with Kaiser normalization.
Eigenvalue, variability percentage, and cumulative percentage using Spearman methods
| Component | Initial Eigenvalues | ||
|---|---|---|---|
| % of variance | % of variance | % of variance | |
| 1 | 11.008 | 35.508 | 35.508 |
| 2 | 5.023 | 16.202 | 51.710 |
| 3 | 3.515 | 11.339 | 63.049 |
| 4 | 2.838 | 9.156 | 72.205 |
| 5 | 1.605 | 5.178 | 77.383 |
| 6 | 1.294 | 4.173 | 81.556 |
| 7 | 1.017 | 3.280 | 84.836 |
| 8 | 0.757 | 2.443 | 87.280 |
| 9 | 0.672 | 2.168 | 89.448 |
| 10 | 0.592 | 1.910 | 91.358 |
| 11 | 0.479 | 1.546 | 92.905 |
| 12 | 0.449 | 1.448 | 94.353 |
| 13 | 0.387 | 1.249 | 95.602 |
| 14 | 0.301 | 0.970 | 96.571 |
| 15 | 0.252 | 0.811 | 97.383 |
| 16 | 0.209 | 0.676 | 98.058 |
| 17 | 0.137 | 0.442 | 98.500 |
| 18 | 0.095 | 0.306 | 98.806 |
| 19 | 0.081 | 0.261 | 99.067 |
| 20 | 0.073 | 0.235 | 99.302 |
| 21 | 0.046 | 0.149 | 99.658 |
| 22 | 0.038 | 0.122 | 99.780 |
| 23 | 0.027 | 0.086 | 99.866 |
| 24 | 0.018 | 0.058 | 99.924 |
| 25 | 0.012 | 0.038 | 99.962 |
| 26 | 0.007 | 0.024 | 99.986 |
| 27 | 0.003 | 0.010 | 99.996 |
| 28 | 0.001 | 0.003 | 99.999 |
| 29 | 0.000 | 0.001 | 100.000 |
| Component | Initial Eigenvalues | ||
|---|---|---|---|
| % of variance | % of variance | % of variance | |
| 1 | 11.008 | 35.508 | 35.508 |
| 2 | 5.023 | 16.202 | 51.710 |
| 3 | 3.515 | 11.339 | 63.049 |
| 4 | 2.838 | 9.156 | 72.205 |
| 5 | 1.605 | 5.178 | 77.383 |
| 6 | 1.294 | 4.173 | 81.556 |
| 7 | 1.017 | 3.280 | 84.836 |
| 8 | 0.757 | 2.443 | 87.280 |
| 9 | 0.672 | 2.168 | 89.448 |
| 10 | 0.592 | 1.910 | 91.358 |
| 11 | 0.479 | 1.546 | 92.905 |
| 12 | 0.449 | 1.448 | 94.353 |
| 13 | 0.387 | 1.249 | 95.602 |
| 14 | 0.301 | 0.970 | 96.571 |
| 15 | 0.252 | 0.811 | 97.383 |
| 16 | 0.209 | 0.676 | 98.058 |
| 17 | 0.137 | 0.442 | 98.500 |
| 18 | 0.095 | 0.306 | 98.806 |
| 19 | 0.081 | 0.261 | 99.067 |
| 20 | 0.073 | 0.235 | 99.302 |
| 21 | 0.046 | 0.149 | 99.658 |
| 22 | 0.038 | 0.122 | 99.780 |
| 23 | 0.027 | 0.086 | 99.866 |
| 24 | 0.018 | 0.058 | 99.924 |
| 25 | 0.012 | 0.038 | 99.962 |
| 26 | 0.007 | 0.024 | 99.986 |
| 27 | 0.003 | 0.010 | 99.996 |
| 28 | 0.001 | 0.003 | 99.999 |
| 29 | 0.000 | 0.001 | 100.000 |
Extraction sums of squared loadings
| Component | Extraction sums of squared loadings | ||
|---|---|---|---|
| % of variance | % of variance | % of variance | |
| 1 | 11.008 | 35.508 | 35.508 |
| 2 | 5.023 | 16.202 | 51.710 |
| 3 | 3.515 | 11.339 | 63.049 |
| 4 | 2.838 | 9.156 | 72.205 |
| 5 | 1.605 | 5.178 | 77.383 |
| 6 | 1.294 | 4.173 | 81.556 |
| 7 | 1.017 | 3.280 | 84.836 |
| Component | Extraction sums of squared loadings | ||
|---|---|---|---|
| % of variance | % of variance | % of variance | |
| 1 | 11.008 | 35.508 | 35.508 |
| 2 | 5.023 | 16.202 | 51.710 |
| 3 | 3.515 | 11.339 | 63.049 |
| 4 | 2.838 | 9.156 | 72.205 |
| 5 | 1.605 | 5.178 | 77.383 |
| 6 | 1.294 | 4.173 | 81.556 |
| 7 | 1.017 | 3.280 | 84.836 |
IHA indices of group II were altered within the 5.882–100% range, as shown in Table 2. From the table, IHA indices indicate slight, low, medium, and high alteration.
CASE STUDY
Jinxi River Profile
Jinxi is a tributary of the Futun River, a minor tributary of the Minjiang River's upper levels. It has a drainage area of 7,201 km2 and is situated between 26° 24′ ∼27° 07′ N and 116° 30′ ∼117° 56′ E. Jinxi originates from the east foot of the Shanling mountains, and the upstream Ningxi flows from Anyuan from southwest to northeast, Zuonalan River, Yanglin River, Datian River, and other rivers until Meikou meets Taining River; the river runs from north-west to south in a deep mountain canyon, passes through Yikou, flows to the southeast after right Nadab Creek at Guanjiang estuary, and has a large tributary Jia. It originates on the right bank and travels eastward. Many short and minor tributaries are incorporated along the route, and the water flows into Futun Creek through Huangtan, Jiangle, and Gaotang before reaching Shunchang. The river has a total length of 253 km and an average gradient of 1.5.
In the Jinxi basin, there are now four power-producing units and nine hydropower plants. Provincial, municipal, and county dispatching are the three types of dispatching. There is a lack of a centralized dispatching command structure. The communication of flood control and power generating dispatching information across various plants and stations, particularly between different owners' power stations, is insufficient and timely. The Jinxi basin comprises 10 runoff power plants and a leading reservoir (Chi Tan Hydropower Station) with yearly regulatory performance. The top reservoir is critical for the basin's economic optimum operation and for improving power plant output at all levels. To fully use the excitement and function of the leading reservoir, it is required to analyse and design a system for the downstream runoff power plant to provide adequate economic compensation to the maximum regulation reservoir while boosting power income.
The minimal ecological flow method of hydropower plants has recently been carefully applied in Fujian Province. The cascade unified dispatching can meet water matching between plants and stations, reducing power generation losses due to ecological flow; at the same time, it can strengthen power generation planning and improve the significant deviation between the power generation plan and the actual situation of the downstream runoff hydropower station group. To some degree, constructing a single operation organization in the Jinxi basin may reduce the conflict between Chitan reservoir water level management and Jinhu tourism's scientific reservoir operation. In the Jinxi basin, there are few legal entities of power production firms. The connection is relatively straightforward, and the basin has a sound operating foundation, all favourable to the project's smooth execution.
Studying river ecological flow is an important research direction in conserving river ecosystems. Ensure the integrity of river ecological functions and safeguard the stability of the river ecosystem by ensuring the ecological flow and change characteristics of rivers. The research of reservoir ecological operation begins with assessing river ecological flow. The meaning and features of river ecological flow, the calculation technique, and the water demand process of river ecological flow should all be clarified. There is no uniform idea or definition of ecological water demand at home or abroad, yet related publications exist in China. The parts and technical standards have no consistent description of ecological water demand.
The entire quantity of water is necessary to sustain the water balance of an ecosystem (including water heat balance, sediment balance, water–salt balance, and water balance) in the appropriate space-time range under a specific ecological aim, which is called ecological water demand. It is concerned with the quality of precipitation resources; in a more limited sense, ecological water demand refers to the amount of runoff water resources that must be supplemented to maintain the ecosystem's normal ecological and environmental functions while meeting a specific ecological goal. Its goal is to improve the quality of runoff water resources. Only the narrowest notion of water resources may be allocated in water resource planning. As a result, studies on the limited idea of ecological water demand for runoff water resources are more realistic.
Chi Tan hydropower station
With a total length of 253 km, an average annual rainfall of 1734.2 mm, and a catchment area of around 7,130 km2, Jinxi is the main tributary of the Futun River in the Minjiang River Basin. Chi Tan Reservoir, finished in 1980, was the first of the Jinxi cascade reservoirs to be developed. Its total storage capacity of 870 million m3 has an imperfect yearly regulation performance. Since the 1990s, the downstream hydropower stations of Liangqian, Dayan, Huangtan, Kongtou, Fancuo, Gaotang, Mowu, and Guiling have all been operational, progressively developing a cascade hydropower development model of ‘one giant with eight little’.
Because the downstream reservoir in the Jinxi basin lacks regulatory capacity, the power station's electricity output is primarily governed by water flowing upstream. The basis of its operating mode is as follows: decrease water abandonment by ultimately generating and pre-evacuating the reservoir according to the hydrological prediction; control high water level operation during other times. It is very susceptible to the outward flow of the Longtou Reservoir Chi Tan power plant due to the downstream power station's weak control performance. Multiple elements, such as power generation, flood discharge, and Chi Tan power plant maintenance, influence the daily power generating plan. It is critical to increase the power production of power plants at all levels to ensure the basin's economic optimum functioning of the Chi Tan reservoir. To fully use the excitement and function of the leading reservoir, it is required to analyse and design a system for the downstream runoff power station to provide adequate economic compensation to the pond reservoir while growing power income.
Study area.
RESULTS AND DISCUSSION
Principal component analysis
The 33 characteristics utilized in IHA are interconnected and may be reduced to representative river indicators using PCA. This work used PCA to determine the smallest number of usual indicators of hydrologic change in the Jinxi River basin. The XLSTAT program was used to do principal component analysis using Pearson and Spearman techniques.
The component plot of rotating space Varimax with Kaiser normalization.
The component plot of rotating space Varimax with Kaiser normalization.
The variance of the principal component axis is directly proportional to its corresponding eigenvalue, and a higher eigenvalue clarifies a higher amount of variation. And component plots on rotated space using the Spearman method. The PCs with eigenvalues greater than 1.0 are used. In this study, the Spearman method considers seven eigenvalues, respectively. The eigenvalues, cumulative percentage, and variability percentage are shown in Table 3.
Retention of meaningful component axes is done based on Kaiser Normalization, whereby axes with eigenvalues greater than 1.0 are taken for further analysis. Seven eigenvalues from the Spearman method show 100% of the total variance in data (Table 4).
Orthogonal rotation is done to select subspaces from the retained PCs. The orthogonal process proposed by Kaiser is used to get factor loadings. For the Spearman method (Tables 5 & 6), F1 is related to the monthly flow of August, September, and October, magnitude and duration of all minimum and maximum flow, base flow, and low pulse; F2 is related to 90 days minimum and all full flow; F3 is related to rise rate and fall rate; F4 is related to monthly flow of February; F5 and F7 are related to date of maximum and minimum flow; F6 show mix loading with no particular dominance.
The KMO MSA has been performed to reduce the number of indicators. Table 7 lists IHA parameters with KMO values greater than 0.8 for the Spearman method. Ten parameters from the Spearman method are retained, with six monthly flows and four indices representing minimum flows in the river. The representative parameters for Jinxi River are the mean monthly flow for January, February, August, September, October, and November and the mean minimum flow for 7 days. These six indicators might be used to assess the hydrological alteration of the Jinxi River basin.
Prominent hydrological indicators based on Spearman methods
| Parameter | Factor loading | PC axis |
|---|---|---|
| Jan | 0.852 | F6 |
| Feb | 0.837 | F6 |
| Aug | 0.877 | F2 |
| Sep | 0.860 | F2 |
| 1-day max | 0.897 | F1 |
| 30-day min | 0.801 | F3 |
| 7-day max | 0.951 | F1 |
| 3-day max | 0.959 | F1 |
| Oct | 0.799 | F2 |
| Nov | 0.794 | F5 |
| Parameter | Factor loading | PC axis |
|---|---|---|
| Jan | 0.852 | F6 |
| Feb | 0.837 | F6 |
| Aug | 0.877 | F2 |
| Sep | 0.860 | F2 |
| 1-day max | 0.897 | F1 |
| 30-day min | 0.801 | F3 |
| 7-day max | 0.951 | F1 |
| 3-day max | 0.959 | F1 |
| Oct | 0.799 | F2 |
| Nov | 0.794 | F5 |
Ecological flow calculation
As explained in earlier chapters, the monthly minimum ecological flow is estimated using the minimum continuous 30d average flow technique, 7Q10 method, and monthly minimum flow calculation method, utilizing a lengthy historical sequence of natural flows from 1982 to 2013. This research uses the most significant value of these three approaches as the minimal flow. The results are shown in Table 8.
Minimum ecological flow calculation results
| Month Station | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
ChiTan hydropower station ( ) | 54.54 | 71.77 | 121.35 | 167.55 | 187.22 | 225.7 | 130.56 | 101.57 | 78.68 | 58.64 | 51.47 | 49.87 |
| Month Station | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ChiTan hydropower station () | 54.54 | 71.77 | 121.35 | 167.55 | 187.22 | 225.7 | 130.56 | 101.57 | 78.68 | 58.64 | 51.47 | 49.87 |
Using the month-by-month frequency calculation approach, the appropriate ecological flow is estimated. The year is separated into three seasons: dry season (October, November, December, January, February), average season (March, April, July, August), and rainy season (September, October, November, December, January, February) (May, June). Then, for the dry season, a guaranteed rate of 90% was employed; for the ordinary season, 70%, and for the rainy year, 50%. Table 9 shows the suitable ecological flow for the Chi Tan reservoir.
Suitable ecological flow calculation results
| Month Station | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
ChiTan hydropower station ( ) | 59.3 | 78.7 | 158.1 | 248.0 | 367.0 | 532.5 | 292.0 | 209.0 | 135.0 | 87.3 | 67.4 | 51.2 |
| Month Station | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ChiTan hydropower station () | 59.3 | 78.7 | 158.1 | 248.0 | 367.0 | 532.5 | 292.0 | 209.0 | 135.0 | 87.3 | 67.4 | 51.2 |
is the minimum ecological flow on the jth day of the ith month, 
is the natural flow on the jth day of the ith month, 
is the average value of the natural flow of the ith month, and 
is the minimum/suitable ecological flow of the ith month. Figure 6 shows the monthly flow downstream of ChiTan hydropower station. 
Monthly flow downstream of ChiTan hydropower station.
The NSGA-II does not focus on the greatest answer for a single goal, but rather offers a variety of optimum and diversified solutions for numerous goals. The Pareto-optimal front of the normal, wet, and dry years are discussed and are shown below:
Normal year:
Wet year
Pareto-optimal solution of the model considering minimum ecological flow (normal year).
Pareto-optimal solution of the model considering minimum ecological flow (normal year).
Pareto-optimal solution of the model considering suitable ecological flow (normal year).
Pareto-optimal solution of the model considering suitable ecological flow (normal year).
Dry year
Pareto-optimal solution of the model considering minimum ecological flow (wet year).
Pareto-optimal solution of the model considering minimum ecological flow (wet year).
Pareto-optimal solution of the model considering suitable ecological flow wet year).
Pareto-optimal solution of the model considering suitable ecological flow wet year).
Pareto-optimal solution of the model considering minimum ecological flow (dry year).
Pareto-optimal solution of the model considering minimum ecological flow (dry year).
Pareto-optimal solution of the model considering suitable ecological flow (dry year).
Pareto-optimal solution of the model considering suitable ecological flow (dry year).
The NSGA-II is very capable of dealing with multipurpose water resource management issues. The suggested model aids decision-makers in finding the Pareto optimum boundary between ecological and human demands by using this optimization approach. The benefits and downsides of hydropower generation must be balanced, and ecological demands must be recognized. As a result, reservoir operations must be coordinated by balancing human and river needs. According to the NGSA-II data, the ecological water scarcity is more than the minimal ecological demand when considering the appropriate demand. This demonstrates that the stronger the environmental protection, the larger the advantage of electricity production that must be sacrificed.
The inquiries in Figure 7–12 and the notion of Pareto solutions within the framework of ecological flow are pivotal to our collective understanding and decision-making: The term ‘total water shortage’ typically refers to the insufficiency in fulfilling all water requirements, encompassing agricultural, industrial, household, and ecological needs. Suppose the overall water shortage depicted in Figure 7–12 includes ecological water scarcity. In that case, it implies that it considers the water deficit needed to sustain ecological flows, including both the minimum and acceptable amounts. Ecological water scarcity refers to insufficient water availability to meet the ecological flow needs for a sustainable and thriving environment. These needs include the water required for fish migration, sediment transport, and maintaining water quality. The calculation for the minimal and appropriate ecological fluxes can be performed independently.
Pareto solution for minimum and suitable ecological flow requirements.
SUMMARY
The hydrologic modification regime has a substantial impact on the river ecosystem as a result of reservoir operation. The primary motivation behind the construction of most reservoirs was economic, which harmed the river ecosystem. Integrating environmental and human demands into reservoir operations and stream management using regime-based approaches is a major challenge for water resources management in the twenty-first century. The primary focus of most academics is to create an operating system that offers economic and ecological benefits. There is still an issue with the operation of the reservoir. The RVA methodology was employed to investigate the hydrological alteration caused by the construction of a reservoir on the ecological conditions of the river downstream, drawing upon the discussions and concepts related to environmental flow. The ChiTan hydropower project in the Jinxi River Basin is being utilized as a subject of analysis. An assessment of the river basin's general condition is provided. The quantification of the hydrological regime's change downstream of the ChiTan hydropower project is determined. Principal component analysis is employed to generate a collection of representative IHA parameters. The effects of hydrologic modification are compared to those of its previous state. The assessment of climate change does not occur on an individual basis. Further investigation is required to distinguish between the impacts of climate change and human-caused causes.
This study aims to clarify the effects of reservoir operation on the modification of hydrological patterns and river ecosystems, explicitly focusing on the impacts of the ChiTan hydroelectric project on the Jinxi River. This research is essential because it focuses on balancing maximizing hydropower generation and preserving ecological well-being. The RVA approach was utilized to examine hydrological alteration. In addition, PCA was used to create representative IHA characteristics, which offers a reliable framework for evaluating alterations. The key findings indicate significant alterations in the hydrological patterns downstream of the ChiTan project compared to its status before it became operational. These encompass changes in the patterns of water movement, which have substantial consequences for water supply and ecological circumstances. This study emphasizes the significance of differentiating between the effects of climate change and those induced by human activity, highlighting the necessity for additional research in this field. Methods for computing monthly minimum ecological flows were formulated, and a reservoir-dispatching model that considers ecological requirements was established. This concept guarantees the fulfilment of environmental flow needs, fostering a more sustainable synergy between hydropower operations and river ecosystems. The hydropower scheduling model was constructed considering the river's ecological requirements. The chosen objective functions were to maximize electricity generation and minimize water deficits. The constraints considered were water balance, reservoir storage capacity, and discharge rates. The NSGA-II algorithm was employed for optimization, successfully balancing these conflicting objectives. This complete approach offers a practical strategy for incorporating ecological factors into hydropower scheduling, ultimately benefiting energy security and environmental preservation.
The PCA technique is also utilized to eliminate statistical redundancy; IHA indicators cannot be wholly reduced using just statistical methods. The monthly minimum ecological flows are calculated using the minimum continuous 30-day average flow techniques, the 7Q10 method, and the monthly minimum flow calculation method. The final value is the sum of the three techniques' results. The month-by-month frequency approach is used to determine the monthly appropriate ecological flow. The monthly flow is translated to daily flow using the formula, and the best reservoir-dispatching model is built utilizing the river's minimal and appropriate ecological flow.
The ideal functioning of a hydropower plant in terms of ecological demands is built using modelling principles. The maximum electricity generation and minimizing of water deficit in the river are chosen as the goal functions. As limitations, water balance, reservoir storage capacity, reservoir discharge, and hydropower production are employed. The multiobjective optimization method NSGA-II is utilized. Finally, the multiobjective hydropower scheduling model is built considering the river's ecological demands – the construction of the NSGA-II algorithm.
Implementation
Optimizing electricity generation involves maximizing the efficiency of hydropower facilities while ensuring compliance with ecological limitations. Ensuring sufficient river flow sustains local ecosystems and communities. Implementing models for sustainable water management involves striking a balance between energy production and ecological requirements to ensure the long-term and responsible use of water resources, sustaining river water levels to ensure the survival of aquatic and riparian species' habitats and ensuring adherence to legal obligations for environmental protection and water consumption. They facilitate informed decision-making by offering policymakers data-driven insights to balance energy needs with ecological conservation effectively and minimizing operational expenses through optimizing resource use and strengthening the approval to conduct operations by showcasing dedication to environmental conservation. Integrating with Smart Grids involves utilizing real-time data and optimization algorithms to enhance the reliability and efficiency of the grid, creating novel technologies and methodologies to save and manage water resources, and creating hydroelectric systems capable of adjusting to fluctuating climatic conditions and severe weather occurrences. They are promoting the shift towards sustainable energy sources by maximizing the effectiveness of hydropower generation. They are advancing the NSGA-II algorithm and other optimization approaches to increase their applicability and integrating hydrology, ecology, and engineering research to build complete models and providing training to professionals on implementing multiobjective optimization techniques for the sustainable management of hydropower and disseminating knowledge to communities regarding the advantages of implementing sustainable water and energy management methods. Optimized hydropower plant management can profoundly influence energy production, environmental sustainability, regulatory compliance, and social benefits. These possibilities underscore the substantial impact of such management.
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.
CONFLICT OF INTEREST
The authors declare there is no conflict.
