River ecological water demand multi-scale dynamic evaluation method considering environmental change

Abstract

The invention discloses a river ecological water demand multi-scale dynamic evaluation method considering environmental changes. Comprising the following steps: acquiring historical runoff data of a target river reach, and taking a sequence minimum value as low water runoff of the period; periodically identifying the low water runoff time sequence, performing correlation analysis, drawing a periodic diagram, selecting three periods with the highest correlation as main periods, and constructing a time periodic function; constructing a candidate model and estimating parameters; selecting an optimal model to calculate the dynamic ecological flow, and calculating the traditional ecological flow based on a time-invariant model; performing high-low judgment on the investigation period to obtain a high-low criterion, and combining the high-low criterion, the dynamic ecological flow and the traditional ecological flow to generate a recommendation threshold value; and respectively executing each step on a plurality of scales of year, season and month, and outputting a recommendation threshold value of the corresponding scale. A dynamic model is combined with a traditional fixed value to generate a recommendation threshold value, so that the scientific frontier of the method is ensured, and the feasibility and robustness of management decision are enhanced.

Description

Technical Field

[0001] This invention relates to the field of ecological flow scheduling decision-making technology, and more specifically to a multi-scale dynamic assessment method for river ecological water demand that takes into account environmental changes. Background Technology

[0002] In recent years, the demand for ecological flow scheduling has continued to grow. Existing ecological flow calculations mostly employ fixed threshold methods and are generally based on the assumption of stationarity. These methods are unable to reflect the non-stationarity caused by climate change and human activities, nor do they adequately characterize the significant differences in runoff at intra-annual seasonal scales, resulting in insufficient targeting of ecological flow targets and weak adaptability to management practices.

[0003] However, the rapid socio-economic development of the basin has simultaneously exacerbated the conflict between domestic, industrial, and ecological water use, leading to problems such as insufficient ecological flow, flow interruption, biodiversity loss, and declining ecosystem service functions in some river sections. Against this backdrop, scientifically defining and effectively guaranteeing ecological water demand within the river channel and maintaining a healthy eco-hydrological relationship within the basin has become a critical issue that urgently needs to be addressed. Existing methods mostly operate on an interannual scale, failing to fully consider the impact of seasonal variations on ecological water demand and process maintenance, thus making it difficult to provide actionable thresholds for seasonal and monthly ecological water allocation.

[0004] Therefore, it is necessary to construct a dynamic ecological flow framework that can change over time and operate at multiple scales such as annual, quarterly, and monthly; at the same time, the dynamic results should be organically linked with traditional fixed ecological flows to form actionable recommended thresholds, so as to improve the feasibility of engineering applications and the level of refinement of management decisions while ensuring scientific validity. Summary of the Invention

[0005] In view of the above problems, the present invention is proposed to provide a multi-scale dynamic assessment method for river ecological water demand that takes into account environmental changes, in order to overcome or at least partially solve the above problems.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] This invention provides a multi-scale dynamic assessment method for river ecological water demand considering environmental changes; including the following steps: S1. Obtain historical runoff data for the target river section, and based on the preset observation period, form a seven-day moving average flow sequence. Use the minimum value of this sequence as the low-water runoff for that period to form a low-water runoff time series. S2. Fourier analysis was used to identify the periodicity of the low-water runoff time series, generating trigonometric function sequences corresponding to different periods. Correlation analysis was performed with the hydrological sequence as the dependent variable and the trigonometric function sequence as the independent variable, and a periodicity diagram was drawn. The three periods with the highest correlation were selected as the main periods to construct a time periodic function. S3. Construct candidate models and estimate parameters within a preset probability distribution parameter set. The candidate models include time-invariant models, trend-varying time-varying models, and periodic time-varying models that embed the time periodic function. S4. Select the optimal model based on the goodness of fit and information criterion, calculate the dynamic ecological flow under the target guarantee rate, and calculate the traditional ecological flow based on the time-invariant model under the target guarantee rate. The traditional ecological flow is a single fixed value within the same observation period. S5. The observation period is divided into abundant and dry periods to obtain the abundant and dry period criteria, and the abundant and dry period criteria, dynamic ecological flow and traditional ecological flow are combined to generate recommended thresholds.

[0008] Furthermore, in step S1, the historical runoff data of the target river section is preprocessed data; The preprocessing process includes missing value imputation, outlier robustness processing, and sequence consistency testing.

[0009] Further, in step S2, the expression for the time periodic function is:

[0010] In the formula, a0, a k b k The expression is:

[0011]

[0012]

[0013] In the formula, a0, a k b k T represents the Fourier coefficients determined by fitting historical sequences. i Let t be the i-th main period, k represent the period number, and t be the time variable.

[0014] Furthermore, in step S3, the specific process of constructing a candidate model and estimating the parameters within the preset probability distribution parameter set includes: A set of multiple distribution parameters is used to broadly capture the probability patterns that the data may follow; For each low-water runoff time series, a time-invariant model, a trend-time-varying model, and a periodic-time-varying model were established and compared. Among them, the time-invariant model assumes that the distribution parameters are constant; Trend-varying time-varying model: assumes that the distribution parameters are polynomial functions of time; Periodic time-varying model: It is assumed that the distribution parameters are linear functions of the time periodic function.

[0015] Furthermore, in step S4, the specific process of calculating dynamic ecological flow includes: By substituting the time period function corresponding to each specific year into the selected optimal model, the probability distribution of that year is obtained; Calculate the quantiles of the probability distribution under the target guarantee rate; The quantiles are calculated year by year to obtain the dynamic ecological flow that changes over time.

[0016] Furthermore, in step S4, the specific process of calculating traditional ecological flow includes: A time-invariant probability model was fitted using the low-water runoff time series to estimate the fixed distribution parameters; The quantiles of the fixed distribution parameters are calculated under the target guarantee rate to obtain the fixed traditional ecological flow values ​​during the observation period.

[0017] Furthermore, in step S5, the wet and dry seasons are determined by calculating the long-term average of the longest period moving average curve in the main cycle as a threshold, and the time period above the threshold is determined as the wet season, and the time period below the threshold is determined as the dry season.

[0018] Furthermore, the recommended threshold satisfies the following condition: When C(P) = dry water, let ,but

[0019] When C(P) = abundant water, let ,but

[0020] In the formula, C(P) is the criterion for determining abundance or scarcity, and A p For the dry year subset, B p For the wet years, t is the time variable, and P is the observation period of the dry runoff time series. This refers to dynamic ecological flow that changes over time. Traditional ecological flows within the same observation period, The threshold is the recommended threshold, and mean is the average value of dynamic ecological flow values ​​that meet the specific year type conditions.

[0021] As can be seen from the above technical solution, compared with the prior art, the present invention discloses a multi-scale dynamic assessment method for river ecological water demand that takes into account environmental changes, which has the following beneficial effects: This application addresses the current need for refined management of river and lake ecological flow, which requires consideration of "water quantity, flow rate, and process," through a multi-scale, year-specific dynamic threshold. This approach enables a more scientific balance between water use for ecological environmental protection and socio-economic development. The recommended thresholds generated by combining dynamic model results with traditional fixed values ​​ensure both the scientific rigor of the methodology and the feasibility and robustness of management decisions. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0023] Figure 1 This is a flowchart of the multi-scale dynamic estimation method for river ecological water demand provided in this embodiment of the invention; Figure 2 This is a Fourier periodicity analysis diagram on an annual scale provided in an embodiment of the present invention; Figure 3 This is an example of an annual flow low value sequence, its periodic fitting curve, and its average value plot on a large periodic scale provided in an embodiment of the present invention. Figure 4 This is a graph showing ecological flow estimation data at different time scales provided in the embodiments of the present invention. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] This invention discloses a multi-scale dynamic assessment method for river ecological water demand that takes into account environmental changes, such as... Figure 1 As shown, it includes the following steps: S1. Obtain historical runoff data for the target river section, and based on the preset observation period, form a seven-day moving average flow sequence. Use the minimum value of this sequence as the low-water runoff for that period to form a low-water runoff time series. S2. Fourier analysis was used to identify the periodicity of the low-water runoff time series, generating trigonometric function sequences corresponding to different periods. Correlation analysis was performed with the hydrological sequence as the dependent variable and the trigonometric function sequence as the independent variable, and a periodicity diagram was drawn. The three periods with the highest correlation were selected as the main periods to construct a time periodic function. S3. Construct candidate models and estimate parameters within a preset probability distribution parameter set. The candidate models include time-invariant models, trend-varying time-varying models, and periodic time-varying models that embed the time periodic function. S4. Select the optimal model based on the goodness of fit and information criterion, calculate the dynamic ecological flow under the target guarantee rate, and calculate the traditional ecological flow based on the time-invariant model under the target guarantee rate. The traditional ecological flow is a single fixed value within the same observation period. S5. The observation period is divided into abundant and dry periods to obtain the abundant and dry period criteria, and the abundant and dry period criteria, dynamic ecological flow and traditional ecological flow are combined to generate recommended thresholds.

[0026] This invention, through its multi-scale, year-specific dynamic thresholds, responds to the current demand for refined management of river and lake ecological flow, which needs to consider "water quantity, flow rate, and process," enabling a more scientific balance between water use for ecological environmental protection and socio-economic development. The recommended thresholds generated by combining dynamic model results with traditional fixed values ​​ensure both the scientific rigor of the method and enhance the feasibility and robustness of management decisions.

[0027] The following is a detailed description of each of the above steps: Step S1: Basic data acquisition and preprocessing; Obtain long-term daily flow observation data from the outlet control hydrological stations of the target basin, typically for no less than 30 years.

[0028] After acquiring the data, it is first preprocessed. The methods used in the preprocessing process include missing value imputation, outlier identification and handling, and sequence consistency testing to ensure data quality.

[0029] Based on the preprocessed observation data described above, the low-water runoff time series Q was calculated. 7min (P,t); Low-water runoff time series Q 7min (P,t) represents the minimum average flow rate over seven consecutive days within the specified observation period P. In this step, the low-water runoff time series Q is calculated using year, month, and day as the preset observation period P, respectively. 7min (P,t).

[0030] Calculate the time series Q of low water runoff 7min The specific process of (P,t) is as follows: 1. Within each preset observation period P, the average flow rate of all consecutive N days is calculated with a step size of one day to obtain the moving average series, where N≥2. Under normal circumstances, the average flow rate of 7 consecutive days is selected within the preset observation period. 2. Take the minimum value of the moving average sequence as the low-water runoff value for the preset observation period P; 3. Calculate the low-water runoff time series year by year, quarter by quarter, and month by month, respectively, and denote them as Q. 7min (year, t), Q 7min (Season, t), Q 7min (month, t).

[0031] In this embodiment, the observation period P, which is measured in years, is called the hydrological year, which is defined as the period from May of the current year to April of the following year. The observation period P, which is measured in seasons and months, is divided according to natural quarters and natural months, as well as according to hydrological characteristic periods such as flood season and non-flood season.

[0032] Step S2: Component identification and periodic function construction; Based on the low-water runoff time series Q calculated in step S1 7min (P,t), identifying the low-water runoff time series Q 7min The non-stationary characteristics, especially the periodicity, in (P,t) are used to construct a quantifiable time periodic function T(t).

[0033] The specific process of constructing the time periodic function is as follows: 1. The low-water runoff time series Q obtained in step S1 7min Preliminary analysis of (P,t) yields the low-water runoff time series Q. 7min The basic nonstationary characteristics of (P,t); 2. Fourier analysis was used to analyze the time series Q of low-water runoff. 7min (P,t) is used for periodic identification; 3. Treat the low-flow runoff time series as a time function f(t), and calculate its relationship with different assumed periods T using Fourier transform. i The correlation coefficient R of the sine and cosine sequences i ; 4. Using the hydrological sequence as the dependent variable and the generated trigonometric function sequence as the independent variable, a correlation analysis was performed, with the period T as the factor. i The horizontal axis is represented by the correlation coefficient R. i Draw a periodic graph with the vertical axis as the ordinate; 5. Select the three periods with the highest correlation coefficients and that pass the statistical test as the main periods, and construct a time periodic function T(t) based on the three main periods.

[0034] The expression for the time periodic function T(t) is:

[0035] a0, a k b k The expression is:

[0036]

[0037]

[0038] a0, a k b k T represents the Fourier coefficients determined by fitting historical sequences. i Let t be the i-th main period, k represent the period number, and t be the time variable.

[0039] In this embodiment, the basic non-stationary characteristics include trends and abrupt change points; the analytical methods used include the Mann-Kendall trend test and the Petittt abrupt change point test. Based on statistical tests, the annual low-water runoff from 1964 to 2021 showed an upward trend of 0.003 m³ / s per decade, with an abrupt change point in 1987. Figure 2 The study showed that the low-water runoff time series has significant periodicity, with main periods of 9.5 years, 15.3 years and 40.9 years.

[0040] Based on the three main characteristic periods identified above, the large-period moving average was calculated ( Figure 3 The average value is approximately 0.103 m³ / s, reflecting the stable evolution trend of annual dry runoff over a long period. Using this average value as a baseline threshold, the study period was divided into wet and dry years. Analysis results show that the basin experienced a significant runoff state transition between 1968 and 2008, shifting from a dry to a wet period, with the turning point occurring around 1988. Since 2009, the basin has entered a new dry year cycle (estimated to be approximately 21 years based on the cycle pattern). These results confirm that the evolution of dry runoff in the basin over the past sixty years has been primarily dominated by a periodic mechanism.

[0041] Step S3: Model construction; The model building process includes: 1. By broadly capturing the probability patterns that data (i.e., historical runoff data of the target river section) may follow through a preset probability distribution set D; 2. For each low-flow sequence, establish a time-invariant model F. fix Trend Time-Varying Model F trendand the periodic time-varying model F cycle The three types of models are compared.

[0042] Time-invariant model F fix Assume the distribution parameter is constant; Trend Time-varying Model F trend Assume the distribution parameter is a polynomial function of time t; Periodic time-varying model F cycle Assume the distribution parameter is a linear function of the time periodic function.

[0043] In this embodiment, the probability distribution set D includes at least five three-parameter distributions: Generalized Gamma (GG), Power Exponential (PE), Box-Cox Cole and Green (BCCG), Normal (NOF) family of distributions, and Exponentially Modified Gaussian (exGAUS); and five two-parameter distributions: Gamma (GA), Gumbel (GU), Lognormal (LO), Weibull (WEI), and Inverse Gamma (IG). Taking the generalized gamma distribution as an example, its location parameter μ and scale parameter... σ Establish a relationship with T(t) through a connection function:

[0044]

[0045] In the formula, α0 and β0 are the intercept terms of the regression model, and α1 and β1 are the regression coefficients corresponding to the location parameter and the scale parameter, respectively.

[0046] Step S4: Calculate dynamic ecological flow and traditional ecological flow; The dynamic ecological flow calculation process is as follows: 1. Selecting the optimal model F based on goodness of fit and information criterion ; 2. Based on the selected optimal model F For each specific time scale (investigation period), the corresponding time covariate is substituted into the model to obtain the probability distribution unique to that time scale (investigation period). 3. Calculate the quantiles of this specific distribution under the target guarantee rate R. ; 4. Calculate annually / quarterly / monthly to obtain a dynamic ecological flow that changes over time. .

[0047] The traditional process for calculating ecological flow is as follows: 1. Based on the time-invariant model F fix A fixed set of distribution parameters is estimated over the entire sequence; 2. Under the same guarantee rate R, calculate the quantile of the guarantee rate R for this fixed distribution to obtain a unique and fixed traditional ecological flow value within the observation period P. .

[0048] In this embodiment, the optimal model is preferably a periodic time-varying model. For the vast majority of time scales, the periodic time-varying model is superior to the time-invariant model and the trend-varying model, as it can better describe the non-stationary characteristics in the data. In the few instances where the time-invariant model and the trend-varying model are superior to the time-invariant model, the differences between them and the periodic time-varying model can be ignored.

[0049] Step S5: Abundance / Sparseness Classification and Recommendation Threshold Generation; The specific process for determining whether an area is abundant or scarce is as follows: Based on the longest period moving average curve in the main period constructed above, the long-term mean of the moving average curve is calculated as the threshold. Years above this threshold are classified as wet years, and years below this threshold are classified as dry years.

[0050] The specific process for generating the recommendation threshold is as follows: Combine dynamic results with traditional results; When a certain observation period P is determined to be a dry year, the recommended threshold is: all dynamic ecological flow values ​​q R In (P,t), those values ​​less than or equal to the traditional fixed value The average of the values. This ensures that a prudent and more reliable lower limit for flow rate is recommended during periods of water shortage.

[0051] When the year is classified as a high-water year, the recommended threshold is: all dynamic ecological flow values ​​q. R In (P,t), those values ​​greater than or equal to the traditional fixed values The average of the values. This allows for setting higher ecological goals to promote ecological restoration when water resources are relatively abundant.

[0052] In this embodiment, as Figure 4As shown, the data are presented on spring, summer, autumn, winter, and year-round scales. Overall, the ecological flow estimation results are highly consistent with the periodic pattern of low flow, showing higher ecological flow estimates in high-flow years and lower estimates in low-flow years. For example, on an annual scale, the improved method estimates relatively low ecological flows in dry years such as 1969–1980 and 2014–2021, and relatively high flows in wet years such as 1964–1968 and 1981–2013. This is largely consistent with the definition of wet and dry years based on dry runoff time series. In contrast, the traditional 7Q10 method, based on the stationarity assumption, is only applicable to years with hydrological conditions close to average (such as 1976 and 1977). It underestimates ecological demand in wet years and overestimates it in dry years, making it difficult to adapt to actual hydrological fluctuations. Perform steps S1-S5 above on three time scales: year, month, and day, and output the recommended threshold for the corresponding scale.

[0053] In this embodiment, recommended ecological flow thresholds (unit: m³ / s) at different time scales were obtained after applying the above method to a certain location. These recommended values ​​are derived from the arithmetic mean of all dynamic ecological flow estimates for the corresponding hydrological year type. Therefore, the recommended values ​​for wet years (e.g., 0.054 m³ / s on an annual scale) are generally higher than those for dry years (0.018 m³ / s). Furthermore, as the time scale becomes more refined (from an annual scale to a flood season scale, and then to a shorter summer scale), the recommended ecological flow values ​​show an increasing trend (e.g., from 0.054 m³ / s to 0.073 m³ / s, and then to 0.086 m³ / s). These multi-scale, year-specific recommended values ​​provide crucial scientific basis for maintaining the integrity of river ecosystems and implementing adaptive water resource management.

[0054] Table 1. Recommended thresholds for ecological flow at different time scales

[0055] The calculated recommended thresholds provide support for drought and flood prevention and control management. Taking the spring of a wet year as an example, the calculated recommended value is 0.115. When the monitoring value exceeds 0.115, it means that the flow is too high and appropriate control of ecological flow is needed. When the monitoring value is lower than the recommended value (0.055) in a dry year, it means that the flow is too low and may not be able to meet the ecological water demand during that period, posing a risk of drought disaster.

[0056] This invention can be applied independently to different time scales such as year, season, and month, and outputs a complete hierarchical ecological flow threshold system.

[0057] Annual-scale output: Provides recommended values ​​for ecological flow throughout the year for macro-control of annual water resource allocation planning in the basin; Output at a seasonal scale: Provides recommended values ​​for spring, summer, autumn and winter, which can guide the seasonal operation of reservoirs; Output on a monthly scale: Provides more refined monthly recommended values, which can be used to generate real-time ecological scheduling instructions on a monthly or even ten-day basis to achieve precise protection.

[0058] This multi-scale, year-specific dynamic threshold responds to the current demand for refined management of river and lake ecological flow, which needs to consider "water quantity, flow rate, and process," and can more scientifically balance the contradiction between water use for ecological environmental protection and socio-economic development. The recommended threshold generated by combining the results of the dynamic model with traditional fixed values ​​ensures both the scientific advancement of the method and enhances the feasibility and robustness of management decisions.

[0059] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0060] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A multi-scale dynamic assessment method for river ecological water demand considering environmental changes, characterized in that, Includes the following steps: S1. Obtain historical runoff data for the target river section, and based on a preset observation period, form an N-day moving average flow sequence. Use the minimum value of the N-day moving average flow sequence as the low-water runoff for that period to form a low-water runoff time series. S2. Fourier analysis was used to identify the periodicity of the low-water runoff time series, generating trigonometric function sequences corresponding to different periods. Correlation analysis was performed with the low-water runoff time series as the dependent variable and the trigonometric function sequences as the independent variables, and a periodicity graph was plotted. The three periods with the highest correlation were selected as the main periods to construct a time periodic function. S3. Construct candidate models and estimate parameters within a preset probability distribution parameter set. The candidate models include time-invariant models, trend-varying time-varying models, and periodic time-varying models that embed the time periodic function. S4. Select the optimal model based on the goodness of fit and information criterion, calculate the dynamic ecological flow under the target guarantee rate, and calculate the traditional ecological flow based on the time-invariant model under the target guarantee rate. The traditional ecological flow is a single fixed value within the same observation period. S5. The observation period is divided into abundant and dry periods to obtain the abundant and dry period criteria, and the abundant and dry period criteria, dynamic ecological flow and traditional ecological flow are combined to generate recommended thresholds.

2. The method for multi-scale dynamic assessment of river ecological water demand considering environmental changes as described in claim 1, characterized in that, In step S1, the historical runoff data of the target river section is the preprocessed data; The preprocessing process includes missing value imputation, outlier robustness processing, and sequence consistency testing.

3. The method for multi-scale dynamic assessment of river ecological water demand considering environmental changes as described in claim 1, characterized in that, In step S2, the expression for the time periodic function is: In the formula, a0, a k b k The expression is: In the formula, a0, a k b k T represents the Fourier coefficients determined by fitting historical sequences. i Let t be the i-th main period, k represent the period number, and t be the time variable.

4. The multi-scale dynamic assessment method for river ecological water demand considering environmental changes as described in claim 1, characterized in that, In step S3, the specific process of constructing a candidate model and estimating parameters within a preset probability distribution parameter set includes: A set of multiple distribution parameters is used to broadly capture the probability patterns that the data may follow; For each low-water runoff time series, a time-invariant model, a trend-time-varying model, and a periodic-time-varying model were established and compared. Among them, the time-invariant model assumes that the distribution parameters are constant; Trend-varying time-varying model: assumes that the distribution parameters are polynomial functions of time; Periodic time-varying model: It is assumed that the distribution parameters are linear functions of the time periodic function.

5. The method for multi-scale dynamic assessment of river ecological water demand considering environmental changes as described in claim 1, characterized in that, In step S4, the specific process of calculating dynamic ecological flow includes: By substituting the time covariate corresponding to each time scale into the selected optimal model, the probability distribution of that year is obtained; Calculate the quantiles of the probability distribution under the target guarantee rate; The quantiles are calculated year by year to obtain the dynamic ecological flow that changes over time.

6. The multi-scale dynamic assessment method for river ecological water demand considering environmental changes as described in claim 1, characterized in that, In step S4, the specific process of calculating traditional ecological flow includes: A time-invariant probability model was fitted using the low-water runoff time series to estimate the fixed distribution parameters; The quantiles of the fixed distribution parameters are calculated under the target guarantee rate to obtain the fixed traditional ecological flow values ​​during the observation period.

7. The method for multi-scale dynamic assessment of river ecological water demand considering environmental changes as described in claim 1, characterized in that, In step S5, the wet and dry seasons are determined by calculating the long-term average of the longest period moving average curve in the main cycle as a threshold. Time periods above the threshold are determined as wet seasons, and time periods below the threshold are determined as dry seasons.

8. The multi-scale dynamic assessment method for river ecological water demand considering environmental changes as described in claim 1, characterized in that, The recommended threshold satisfies the following conditions: When C(P) = dry water, let ,but When C(P) = abundant water, let ,but In the formula, C(P) is the criterion for determining abundance or scarcity, and A p For the dry year subset, B p For the wet years, t is the time variable, and P is the observation period of the dry runoff time series. This refers to dynamic ecological flow that changes over time. Traditional ecological flows within the same observation period The threshold is the recommended threshold, and mean is the average value of dynamic ecological flow values ​​that meet the specific year type conditions.

Images