Flood season anti-dry risk assessment method based on composite runoff index

Abstract

The invention belongs to the technical field of flood and drought disaster recognition and evaluation, and discloses a flood season anti-dry risk evaluation method based on a composite runoff index, and the method comprises the steps: obtaining geographic information data, hydro meteorological data and hydraulic engineering data of a target region; setting a drought threshold in combination with urban water supply requirements to select low-flow drought events in historical flood seasons; optimal probability distribution is selected, and marginal distribution characteristics of drought flow and duration of the low-flow drought event in the flood season are analyzed; selecting a Copula function, establishing joint distribution of drought flow and duration of the low-flow drought event in the flood season, and constructing a composite runoff index of the low-flow drought event in the flood season; and in consideration of different discharge scenes of future climate change, based on a drainage basin distributed hydrological model prediction result, analyzing occurrence risk response characteristics of future flood season anti-flood events by adopting a composite runoff index. The method can accurately evaluate the sudden rising characteristic of flood season anti-dry and the dry water characteristic of the flood season.

Description

Technical Field

[0001] The invention relates to the technical field of water and drought disaster identification and assessment, and in particular to a flood season drought risk assessment method based on a composite runoff index. Background Art

[0002] Drought disasters are the most common natural disasters in the world. As global climate change continues to warm, the intensity and frequency of extreme drought events are increasing. In particular, the emergence of drought events during the flood season poses a huge challenge to flood control and drought relief. It not only brings great uncertainty to flood control, but also threatens the water supply security for economic and social development. The identification and assessment of drought events during the flood season is an important basis for quantifying the level of water and drought disasters and analyzing the impact of drought disasters during the flood season. It is of great significance for water and drought disaster forecasting, response and disposal, and water supply security.

[0003] According to literature research, in order to quantitatively evaluate the characteristics of hydrological drought, commonly used methods include surface water supply index, Palmer hydrological drought index, standardized runoff index, runoff drought index, etc. Taking the most commonly used standardized runoff index as an example, the optimal probability distribution of runoff is generally determined, and then the standardized runoff index is obtained by normal standardization to analyze drought characteristics. Its time scale is generally monthly, seasonal or annual. This coarse time resolution often makes it difficult to capture the sudden characteristics of flood season droughts, and at the same time confuses the dry statistical characteristics that originally belong to the non-flood season into the drought characteristics of the flood season, thereby affecting the prediction ability of flood season drought events. In addition, in the actual drought relief process, since urban water supply infrastructure is often designed based on a certain guarantee rate standard, long-term droughts will put significant pressure on the degree of water supply security. Therefore, it is necessary to focus on the drought flow corresponding to the low water level and its duration and conduct a joint analysis to assess the risk of water supply security. However, there is still a lack of joint characteristic analysis of drought flow and duration of flood season drought events.

[0004] Based on this, it is necessary to further combine the urban water supply requirements, consider the characteristics and laws of the elements of flood season drought events, and improve the design of identification and assessment methods for flood season drought events. Summary of the invention

[0005] In view of the above-mentioned shortcomings of the prior art, the purpose of the present invention is to provide a flood season drought risk assessment method based on a composite runoff index by fully considering the joint characteristics of drought flow and drought duration during the flood season.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] The present invention provides a flood season drought risk assessment method based on a composite runoff index, the method comprising:

[0008] Obtain geographic information data, hydrological and meteorological data, and water conservancy project data of the target area;

[0009] Based on geographic information data, hydrological and meteorological data, water conservancy project data and urban water supply demand, drought thresholds are set to select historical flood season low-flow drought events;

[0010] Select the optimal probability distribution to analyze the marginal distribution characteristics of drought flow and duration of low-flow drought events during the flood season;

[0011] Select the Copula function, establish the joint distribution of drought flow and duration of low-flow drought events during the flood season, and construct the composite runoff index of low-flow drought events during the flood season;

[0012] A distributed hydrological model of the watershed is constructed, and different emission scenarios of future climate change are considered. Based on the prediction results of the distributed hydrological model of the watershed, a composite runoff index is used to analyze the risk response characteristics of future flood season drought events.

[0013] Preferably, the geographic information data of the target area include digital elevation terrain data, river system and hydrometeorological station vector data of the target area; the hydrometeorological data of the target area include long series water level and flow observation data of the hydrological stations in the target area, long series meteorological observation data of the meteorological stations, and long series climate forecast data for different emission scenarios in the future; the water conservancy project data of the target area include spatial distribution vector data and project parameters of the water intake projects in the target area.

[0014] Preferably, based on geographic information data, hydrological and meteorological data, water conservancy project data and urban water supply demand, a drought threshold is set to select historical flood season low-flow drought events, including:

[0015] When the water level of the water intake project and the water level-flow relationship of the water intake river section are known, the drought flow threshold is determined through the water level-flow relationship curve;

[0016] In the absence of the above engineering parameters, the daily runoff process sequence of the water intake section during the flood season was selected, and the generalized extreme value distribution was used for fitting. The flow value under the design frequency condition was selected as the drought threshold according to the urban water supply standard. The distribution function of the generalized extreme value distribution is as follows:

[0017]

[0018] Where F(x) is the distribution function, x is a random variable, μ, ξ, σ are the location parameter, scale parameter and shape parameter respectively, and exp is the empirical function;

[0019] On the basis of setting drought threshold, using historical flood season runoff data, based on the run theory, low-flow drought events in the flood season were selected, the flow and duration of each low-flow drought event were analyzed and calculated, and the drought flow and drought duration series data of low-flow drought events in the flood season were obtained.

[0020] Preferably, the optimal probability distribution is selected to analyze the marginal distribution characteristics of drought flow and duration of low-flow drought events during the flood season, including:

[0021] According to the drought flow and drought duration sequence data of low-flow drought events in the flood season, a univariate probability distribution is used as an alternative distribution type, and the Kolmogorov-Smirnov test, the Bayesian information criterion and the Akaike information criterion are used as evaluation criteria to select the marginal distribution function that best fits the drought flow and duration of low-flow drought events in the flood season; wherein the multiple univariate probability distributions include gamma distribution, Rayleigh distribution, normal distribution, exponential distribution, lognormal distribution, Weibull distribution, extreme value distribution, generalized Pareto distribution, generalized extreme value distribution, log-logistic distribution, logistic distribution and Pearson type III distribution;

[0022] Based on the optimal fitting marginal distribution function of drought flow and duration of low-flow drought events in flood season, the marginal distribution characteristics of drought flow and duration of low-flow drought events in flood season are analyzed.

[0023] Preferably, a Copula function is selected to establish the joint distribution of drought flow and duration of low-flow drought events during the flood season, including:

[0024] Compare multiple two-dimensional Copula functions, use the Bayesian information criterion and Akaike information criterion to evaluate the fitting effect of each Copula function, and select the optimal Copula function to establish the joint distribution function of drought flow and duration of low-flow drought events in flood season. The joint distribution function is expressed as:

[0025]

[0026] In the formula, C(u,v) is the optimal Copula function, x1 and x2 are the low flow drought flow and duration values ​​in the flood season, u and v are the distribution functions of drought flow and duration, and F1 and F2 are the distribution function expressions of drought flow and duration, respectively. are the inverse function expressions of drought flow and duration distribution functions respectively, and H is the function expression of joint distribution.

[0027] Preferably, constructing a composite runoff index for low-flow drought events during the flood season includes:

[0028] According to the joint distribution function, the joint probability sequence of drought flow and drought duration of historical flood season low flow events is calculated, and the sequence is normalized to obtain the corresponding composite runoff index; the calculation process is as follows:

[0029]

[0030] Where, CSI is the composite runoff index, P is the joint probability, S is the sign function, when P>0.5, S is 1.0, when P≤0.5, S is -1.0, t is the intermediate variable, c0, c1, c2, d1, d2, d3 are all constants, c0=2.515517, c1=0.802853, c2=0.010328, d1=1.432788, d2=0.189269, d3=0.001308;

[0031] The flood season drought event levels corresponding to the composite runoff index CSI are divided as follows:

[0032] When CFI is less than -2.0, the level of drought event during the flood season is extreme drought;

[0033] When -2.0≤CSI<-1.5, the level of drought event during the flood season is severe drought;

[0034] When -1.5≤CSI<-1.0, the level of drought event during the flood season is moderate drought;

[0035] When -1.0<CSI≤0.0, the level of drought event during the flood season is mild drought.

[0036] Preferably, a basin distributed hydrological model is constructed, different emission scenarios of future climate change are considered, and based on the prediction results of the basin distributed hydrological model, a composite runoff index is used to analyze the risk response characteristics of future flood season drought events, including:

[0037] Construct a distributed hydrological model of the target area, use digital elevation terrain data, river system data and other data to grid the target area, drive the distributed hydrological model of the basin based on historical meteorological data to simulate runoff changes, and use historical hydrological data to calibrate the model;

[0038] Select future low-emission, medium-emission and high-emission scenarios, and use the quantile mapping method to downscale the meteorological elements output by the model based on historical measured data to obtain long series data of future precipitation, temperature, wind speed, humidity and radiation predicted by different climate models under low-emission, medium-emission and high-emission scenarios;

[0039] The climate prediction data output by different climate models under different emission scenarios in the future are used to drive the calibrated basin distributed hydrological model to predict the future runoff evolution process of the basin. The constructed composite runoff index is used to analyze the risk response characteristics of future flood season drought events. The calculation formula is as follows:

[0040]

[0041] In the formula, R d is the flood season reverse drought risk, m is the total number of flood season low flow drought events, N is the year, CFI i is the composite runoff index value of the low-flow drought event in the i-th flood season, CSI L 、CSI U It is the upper and lower limits of the drought composite runoff index corresponding to the level of flood season drought events.

[0042] Through the above technical solution, the present invention has the following beneficial effects:

[0043] 1. The present invention fully considers the urban water supply demand and constructs a composite runoff index that couples the drought flow and duration of low-flow drought events during the flood season. It can more accurately evaluate the sudden characteristics of flood season drought and the characteristics of flood season drought, and can provide a more scientific basic support for the early warning, forecasting, response and impact assessment of flood season drought disasters.

[0044] 2. The nonlinear flood season low-flow drought composite runoff index constructed by the present invention based on the Copula function overcomes the subjective problem of linear weighting of previous multivariate indicators, and can more objectively reflect the risk response characteristics of flood season drought under different emission scenarios of future climate change, providing scientific and technological support for ensuring the water supply security needs of sustainable development of urban economy and society. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0046] Figure 1 The present invention is a flowchart of a method for assessing flood season drought risk based on a composite runoff index according to an embodiment of the present invention.

[0047] Figure 2 This is a characteristic diagram of changes in the composite runoff index of the target area during the historical period according to an embodiment of the present invention.

[0048] Figure 3This is a risk response characteristic diagram of future flood season drought events in the target area according to an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The following examples are only for better explanation of the present invention, but the content of the present invention is not limited to the application in the examples. Therefore, those skilled in the art can make non-essential improvements and adjustments to the implementation scheme according to the above content of the invention and apply them to other examples, which are still within the protection scope of the present invention.

[0050] The present invention will now be further described with reference to the accompanying drawings.

[0051] Figure 1 Flow chart of a flood season drought risk assessment method based on a composite runoff index according to an embodiment of the present invention. The present invention provides a flood season drought risk assessment method based on a composite runoff index, such as Figure 1 As shown, the method comprises the following steps:

[0052] Step 1: Obtain geographic information data, hydrological and meteorological data, and water conservancy project data of the target area.

[0053] In this embodiment, the geographic information data collected and organized in the target area include digital elevation terrain data, river system and hydrological and meteorological station vector data; the hydrological and meteorological data include the long series of water level and flow observation data of the main hydrological stations in the target area, the long series of meteorological observation data of the meteorological stations, and the long series of climate forecast data for different emission scenarios in the future; the water conservancy project data include the spatial distribution vector data of the main water intake projects in the target area and project parameters.

[0054] In one exemplary embodiment, a city in the middle reaches of the Yangtze River is taken as an example of the target area, and basic geographic information data above the section of the city in the middle reaches of the Yangtze River are collected and organized, including 30m resolution raster data, Yangtze River river system vector data, and spatial distribution vector data of hydrological and meteorological stations; the hydrological and meteorological data include water level and flow observation data of the hydrological station series of the city section of the Yangtze River main stream, daily data such as precipitation, temperature, wind speed, humidity, and sunshine time of the meteorological station series, and daily data such as precipitation, temperature, wind speed, humidity, and radiation predicted under different emission scenarios in the future output by the climate model; the water conservancy project data include spatial distribution vector data of the main water intake projects of the city in the middle reaches of the Yangtze River, and the parameters of the water conservancy projects include the water intake level of the water intake projects.

[0055] Step 2: Based on geographic information data, hydrological and meteorological data, water conservancy project data, and urban water supply demand, drought thresholds are set to select historical flood season low-flow drought events.

[0056] In this embodiment, the drought flow threshold is determined according to the urban water supply safety demand, which can be determined by the following method: when the water level of the water intake project and the water level and flow relationship of the water intake river section are known, the drought flow threshold is determined by the water level and flow relationship curve; in the absence of the above engineering parameters, the flood season daily scale runoff process sequence of the water intake section is selected, and the generalized extreme value distribution (Generalized Extreme Value distribution, GEV) is used for fitting, and the flow value under the design frequency condition is selected as the drought threshold according to the urban water supply standard, and its distribution function is as follows:

[0057]

[0058] Where F(x) is the distribution function, x is a random variable, μ, ξ, σ are the location parameter, scale parameter, and shape parameter, respectively.

[0059] On the basis of setting drought threshold, using historical flood season runoff data, based on the run theory, low-flow drought events in the flood season were selected, the flow and duration of each low-flow drought event were analyzed and calculated, and the drought flow and drought duration series data of low-flow drought events in the flood season were obtained.

[0060] In an exemplary embodiment, based on the historical long series of daily flow data from 1960 to 2018 at a control section of a city in the middle reaches of the Yangtze River, a daily-scale runoff process sequence during the flood season was selected, and then the GEV model was used for distribution fitting to analyze the characteristics of historical drought changes and calculate the five-year return flow as the drought threshold. According to the drought threshold, the historical daily runoff flow data was analyzed using the run theory to obtain the flood season low-flow drought events since 1960-2018, and the time series data of flow and duration were calculated.

[0061] Step 3: Select the optimal probability distribution and analyze the marginal distribution characteristics of drought flow and duration of low-flow drought events during the flood season.

[0062] In this embodiment, according to the drought flow and duration series determined by the low-flow drought event in the flood season selected in step 2, the univariate probability distribution widely used in hydrological frequency analysis is adopted as the alternative distribution type, including gamma distribution (Gamma), Rayleigh distribution (Rayleigh), normal distribution (Normal), exponential distribution (Exponential), lognormal distribution (Lognormal), Weibull distribution (Weibull), extreme value distribution (Extreme Value), generalized Pareto distribution (Generalized Pareto Distribution, GPD), generalized extreme value distribution (Generalized Extreme Value, GEV), log-logistic distribution (Log-logistic), logistic distribution (Logistic), Pearson III type distribution, etc., and the Kolmogorov-Smirnov test, Bayesian Information Criterion (Bayesian Information Criterion, BIC) and Akaike Information Criterion (AIC) are used as evaluation criteria to select the marginal distribution function that best fits the drought flow and duration of the low-flow drought event in the flood season.

[0063] In an exemplary embodiment, probability distribution fitting is performed on the drought flow and drought duration of low-flow drought events during the flood season in a water intake project control section of a city in the middle reaches of the Yangtze River from 1960 to 2018, and the optimal marginal distribution functions are obtained: the optimal marginal distribution of drought duration is the generalized Pareto distribution (GPD), and the optimal marginal distribution of drought flow is the generalized extreme value distribution (GEV).

[0064] Step 4: Select the Copula function, establish the joint distribution of drought flow and duration of low-flow drought events during the flood season, and construct a composite runoff index for low-flow drought events during the flood season.

[0065] After obtaining the marginal distribution functions of drought flow and duration of low-flow drought events in the flood season in step 3, the Copula function is further used to construct its joint distribution. By comparing various two-dimensional Copula functions including Gaussian Copula, Student t Copula, Clayton Copula, Gumbel Copula, Frank Copula, etc., the Bayesian Information Criterion (BIC) and Akaike Information Criterion (AlC) are used to evaluate the fitting effect of each Copula function, and the optimal Copula model is selected to establish the joint distribution function of drought flow and duration of low-flow drought events in the flood season. The distribution function is as follows:

[0066]

[0067] In the formula, C(u,v) is the Copula function, x1,x2 are the low flow drought flow and duration values ​​during the flood season, u,v are the distribution functions of drought flow and duration, F1,F2 are the distribution function expressions of drought flow and duration, are the inverse function expressions of drought flow and duration distribution functions respectively, and H is the function expression of joint distribution.

[0068] According to the joint distribution function, the joint probability sequence of drought flow and drought duration of historical low flow events in flood season is calculated, and the sequence is normalized to obtain the corresponding composite streamflow index (CSI).

[0069]

[0070] In the formula, CSI is the composite runoff index, P is the joint probability, S is the sign function, when P>0.5, S is 1.0, when P≤0.5, S is -1.0, t is the intermediate variable, and the other parameters are c0=2.515517, c1=0.802853, c2=0.010328, d1=1.432788, d2=0.189269, d3=0.001308. The CSI index can be used to classify the flood season drought event level as follows: CFI<-2.0 is extreme drought, -2.0≤CSI<-1.5 is severe drought, -1.5≤CSI<-1.0 is moderate drought, and -1.0<CSI≤0.0 is light drought.

[0071] In an exemplary embodiment, Gaussian Copula, Student t Copula, Clayton Copula, Gumbel Copula, and Frank Copula are used to fit the joint distribution respectively, and the optimal Copula function is selected as Frank Copula from the above two-dimensional Copula functions using the AIC and BIC criteria. According to the preferred Copula function, the joint probability sequence of low-flow drought flow and drought duration in the flood season is calculated, and standardized to obtain the historical composite runoff index sequence. The results can clearly reflect the historical flood season anti-drought and severe drought events in 1972, 1977, 1984, 2011, and 2018, see Figure 2 .

[0072] Step 5: Construct a distributed hydrological model of the watershed, consider different emission scenarios of future climate change, and use a composite runoff index to analyze the risk response characteristics of future flood season drought events based on the prediction results of the distributed hydrological model of the watershed.

[0073] In this embodiment, first, a distributed hydrological model of the target area basin is constructed, and the target area is gridded using digital elevation terrain data, river system data, and other data. The distributed hydrological model is driven based on historical meteorological data to simulate runoff changes, and the model is calibrated using historical hydrological data; secondly, different future emission scenarios are selected, including low-emission, medium-emission, and high-emission scenarios. According to historical measured data, the quantile mapping method is used to downscale the meteorological elements output by the model to obtain long series data such as future precipitation, temperature, wind speed, humidity, and radiation predicted by different climate models under low-emission, medium-emission, and high-emission scenarios; thirdly, the climate prediction data output by different climate models under different future emission scenarios are used to drive the calibrated distributed hydrological model of the basin to predict the future runoff evolution process of the basin, and the composite runoff index constructed in step 4 is used to analyze the risk response characteristics of future flood season anti-drought events. The calculation formula is as follows:

[0074]

[0075] In the formula, R d is the flood season reverse drought risk, m is the total number of flood season low flow drought events, N is the year, CFI i is the composite runoff index value of the low-flow drought event in the i-th flood season, CSI L 、CSI U These are the upper and lower limits of the composite runoff index for this level of drought.

[0076] In an exemplary embodiment, by constructing a distributed time-varying gain hydrological model of the target area, an optimization algorithm is used to calibrate the model parameters based on the historical hydrological observation data of the control section hydrological stations. On this basis, the climate prediction results of four future global climate models including ACCESS-CM2, EC-Earth3, MRI-ESM2-0, and NorESM2-LM are used and downscaled to drive the calibrated time-varying gain hydrological model to predict different future scenarios including low emission, medium emission and high emission scenarios, and predict the risk response characteristics of future flood season drought events. The risk response characteristics of flood season drought events in the target area in the middle of this century (horizontal year 1) and the end of this century (horizontal year 2) are shown in Figure 3 As shown in the results, it can be seen that except for the mild drought in the low emission scenario and the extreme drought in the medium emission scenario, the anti-drought during the flood season shows a decreasing trend, and the extreme drought will increase significantly under the medium emission scenario by the end of this century.

[0077] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some or all of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and should all be included in the scope of the claims and specification of the present invention.

Claims

1. A flood season drought risk assessment method based on a composite runoff index, characterized in that: The method comprises: Obtain geographic information data, hydrological and meteorological data, and water conservancy project data of the target area; Based on geographic information data, hydrological and meteorological data, water conservancy project data and urban water supply demand, drought thresholds are set to select historical flood season low-flow drought events; Select the optimal probability distribution to analyze the marginal distribution characteristics of drought flow and duration of low-flow drought events during the flood season; Select the Copula function, establish the joint distribution of drought flow and duration of low-flow drought events during the flood season, and construct the composite runoff index of low-flow drought events during the flood season; A distributed hydrological model of the watershed is constructed, and different emission scenarios of future climate change are considered. Based on the prediction results of the distributed hydrological model of the watershed, a composite runoff index is used to analyze the risk response characteristics of future flood season drought events.

2. The flood season drought risk assessment method based on composite runoff index according to claim 1 is characterized in that: The geographic information data of the target area include the digital elevation terrain data, river system and hydrological and meteorological station vector data of the target area; the hydrological and meteorological data of the target area include the long series of water level and flow observation data of the hydrological stations in the target area, the long series of meteorological observation data of the meteorological stations, and the long series of climate forecast data for different emission scenarios in the future; the water conservancy project data of the target area include the spatial distribution vector data of the water intake projects in the target area and the project parameters.

3. The flood season drought risk assessment method based on composite runoff index according to claim 1 is characterized in that: Based on geographic information data, hydrological and meteorological data, water conservancy project data and urban water supply demand, drought thresholds are set to select historical flood season low-flow drought events, including: When the water level of the water intake project and the water level-flow relationship of the water intake river section are known, the drought flow threshold is determined through the water level-flow relationship curve; In the absence of the above engineering parameters, the daily runoff process sequence of the water intake section during the flood season was selected, and the generalized extreme value distribution was used for fitting. The flow value under the design frequency condition was selected as the drought threshold according to the urban water supply standard. The distribution function of the generalized extreme value distribution is as follows: Where F(x) is the distribution function, x is a random variable, μ, ξ, σ are the location parameter, scale parameter and shape parameter respectively, and exp is the empirical function; On the basis of setting drought threshold, using historical flood season runoff data, based on the run theory, low-flow drought events in the flood season were selected, the flow and duration of each low-flow drought event were analyzed and calculated, and the drought flow and drought duration series data of low-flow drought events in the flood season were obtained.

4. The flood season drought risk assessment method based on composite runoff index according to claim 3 is characterized in that: Select the optimal probability distribution to analyze the marginal distribution characteristics of drought flow and duration of low-flow drought events during the flood season, including: According to the drought flow and drought duration sequence data of low-flow drought events in the flood season, a univariate probability distribution is used as an alternative distribution type, and the Kolmogorov-Smirnov test, the Bayesian information criterion and the Akaike information criterion are used as evaluation criteria to select the marginal distribution function that best fits the drought flow and duration of low-flow drought events in the flood season; wherein the multiple univariate probability distributions include gamma distribution, Rayleigh distribution, normal distribution, exponential distribution, lognormal distribution, Weibull distribution, extreme value distribution, generalized Pareto distribution, generalized extreme value distribution, log-logistic distribution, logistic distribution and Pearson type III distribution; Based on the optimal fitting marginal distribution function of drought flow and duration of low-flow drought events in flood season, the marginal distribution characteristics of drought flow and duration of low-flow drought events in flood season are analyzed.

5. The flood season drought risk assessment method based on composite runoff index according to claim 4 is characterized in that: Select the Copula function to establish the joint distribution of drought flow and duration of low-flow drought events during the flood season, including: Compare multiple two-dimensional Copula functions, use the Bayesian information criterion and Akaike information criterion to evaluate the fitting effect of each Copula function, and select the optimal Copula function to establish the joint distribution function of drought flow and duration of low-flow drought events in flood season. The joint distribution function is expressed as: Where C(u,v) is the optimal Copula function, x1 and x2 are the low flow drought flow and duration values ​​during the flood season, u and v are the distribution functions of drought flow and duration, F1 and F2 are the distribution function expressions of drought flow and duration, and F1 is the distribution function expression of drought flow and duration. -1 , are the inverse function expressions of drought flow and duration distribution functions respectively, and H is the function expression of joint distribution.

6. The flood season drought risk assessment method based on composite runoff index according to claim 5 is characterized in that: Construct a composite runoff index for low-flow drought events during the flood season, including: According to the joint distribution function, the joint probability sequence of drought flow and drought duration of historical flood season low flow events is calculated, and the sequence is normalized to obtain the corresponding composite runoff index; the calculation process is as follows: Where, CSI is the composite runoff index, P is the joint probability, S is the sign function, when P>0.5, S is 1.0, when P≤0.5, S is -1.0, t is the intermediate variable, c0, c1, c2, d1, d2, d3 are all constants, c0=2.515517, c1=0.802853, c2=0.010328, d1=1.432788, d2=0.189269, d3=0.001308; The flood season drought event levels corresponding to the composite runoff index CSI are divided as follows: When CFI is less than -2.0, the level of drought event during the flood season is extreme drought; When -2.0≤CSI<-1.5, the level of drought event during the flood season is severe drought; When -1.5≤CSI<-1.0, the level of drought event during the flood season is moderate drought; When -1.0<CSI≤0.0, the level of drought event during the flood season is mild drought.

7. The flood season drought risk assessment method based on composite runoff index according to claim 6 is characterized in that: A distributed hydrological model of the basin is constructed, and different emission scenarios of future climate change are considered. Based on the prediction results of the distributed hydrological model of the basin, a composite runoff index is used to analyze the risk response characteristics of future flood season drought events, including: Construct a distributed hydrological model of the target area, use digital elevation terrain data, river system data and other data to grid the target area, drive the distributed hydrological model of the basin based on historical meteorological data to simulate runoff changes, and use historical hydrological data to calibrate the model; Select future low-emission, medium-emission and high-emission scenarios, and use the quantile mapping method to downscale the meteorological elements output by the model based on historical measured data to obtain long series data of future precipitation, temperature, wind speed, humidity and radiation predicted by different climate models under low-emission, medium-emission and high-emission scenarios; The climate prediction data output by different climate models under different emission scenarios in the future are used to drive the calibrated basin distributed hydrological model to predict the future runoff evolution process of the basin. The constructed composite runoff index is used to analyze the risk response characteristics of future flood season drought events. The calculation formula is as follows: In the formula, R d is the flood season reverse drought risk, m is the total number of flood season low flow drought events, N is the year, CFI i is the composite runoff index value of the low-flow drought event in the i-th flood season, CSI L 、CSI U It is the upper and lower limits of the drought composite runoff index corresponding to the level of flood season drought events.

Images