A natural-social coupling watershed hydrological drought assessment method and system

Abstract

The application discloses a kind of natural society coupling's watershed hydrology drought assessment method and system, by collecting the precipitation observation data and runoff observation data of target watershed, carry out control section precipitation runoff relationship variation diagnosis;Calculate drought index, including meteorological drought index and hydrological drought index considering the influence of reservoir scheduling;Calculate the drought threshold based on natural society coupling, including standardized meteorological drought threshold and standardized hydrological drought threshold;Based on the results of control section precipitation runoff relationship variation diagnosis, combined with drought index, calculate the response time of hydrological drought to meteorological drought, according to the drought threshold based on natural society coupling, based on Copula function estimation hydrological drought occurrence conditional probability and propagation threshold, based on the conditional probability and propagation threshold of hydrological drought occurrence assesses watershed drought occurrence risk, can accurately depict artificial control watershed meteorological drought triggers the dynamic risk of hydrological drought.

Description

Technical Field

[0001] This invention relates to the field of hydrological drought assessment technology, and more particularly to a natural A socially coupled watershed hydrological drought assessment method and system. Background Technology

[0002] Drought is a natural phenomenon caused by a persistent water deficit due to abnormal water cycle. It spreads through interconnected land-atmosphere systems and water cycle processes, forming four types of drought at different stages of the water cycle: meteorological, hydrological, agricultural, and socio-economic. Drought develops slowly, usually beginning with abnormally low precipitation (meteorological drought), and the resulting reduction in river runoff and lake / reservoir storage (hydrological drought) is the main form of drought-induced disaster.

[0003] Although human societal activities are primarily affected by hydrological drought, meteorological drought is a precursor to hydrological drought, and there is a close intrinsic link between the characteristic variables of the two. Furthermore, long-sequence, high-precision, and high spatiotemporal resolution meteorological data are readily available, and meteorological drought research is relatively mature. The meteorological drought indices such as the Standardized Precipitation Index (SPI) and the Palmer Drought Severity Index (PDSI) are mainly used in drought monitoring operational systems for real-time drought monitoring. Utilizing meteorological drought characteristic variables to assess hydrological drought is a common method for drought management in various countries.

[0004] However, influenced by intensive human activities, the driving forces, cycle structure, and cycle parameters of the watershed's water cycle exhibit a dualistic evolutionary pattern. Therefore, the occurrence and development of drought are influenced by "natural..." Driven by a dual coupling of "social" factors, and jointly controlled by multiple scale factors such as regional climate, hydrological conditions, and human activities, hydrological drought is characterized by numerous driving factors and time-varying contributions. Constructing reservoirs to regulate the spatiotemporal distribution of river runoff as needed is currently the main measure to address drought. However, this also interferes with the formation and development of hydrological drought, thereby altering the pattern of meteorological drought triggering hydrological drought and rendering existing knowledge of hydrological drought patterns established under natural conditions inapplicable. Therefore, exploring the natural... Socially coupled watershed hydrological drought assessment is of great significance for building a scientific and effective drought monitoring and early warning system and realizing the shift in drought relief and disaster reduction from focusing on "post-disaster relief" to focusing on "pre-disaster prevention".

[0005] Current methods for assessing drought in watersheds have three significant shortcomings: 1) Drought thresholds are either fixed as constants from a "nature-centered" perspective (such as the 0 threshold of the standardized index) or defined as empirical percentiles from a "social-centered" perspective. Both of these approaches focus solely on the natural or social dimension of drought, neglecting the fact that drought is a natural phenomenon. 1) The dual nature of socially coupled disasters; 2) Existing studies mostly focus on the response time of hydrological drought to meteorological drought, with insufficient attention to the probability and threshold of a given meteorological drought triggering hydrological drought, and very little consideration of the impact of runoff inconsistency caused by human activities such as reservoir scheduling on the propagation characteristics of meteorological to hydrological drought and the parameters of the Copula correlation function; 3) Existing probabilistic models based on the Copula correlation function only approximate the anthropogenic inconsistency by introducing a drought index modified by covariates, failing to characterize the dynamic risk of meteorological drought triggering hydrological drought in the watershed from a mechanistic perspective, resulting in assessment results that are difficult to truly reflect the actual drought risk situation. Summary of the Invention

[0006] The purpose of this invention is to provide a natural A socially coupled watershed hydrological drought assessment method and system is proposed to address the current watershed hydrological drought assessment methods that one-sidedly focus on the single natural or social dimension of drought in setting drought thresholds, neglecting drought as a natural phenomenon. The dual nature of socially coupled disasters, coupled with a lack of consideration for the impact of runoff inconsistency caused by human activities such as reservoir management on the propagation characteristics of meteorological to hydrological drought and the parameters of Copula-related functions, fails to characterize the dynamic risk of meteorological drought triggering hydrological drought in a watershed from a mechanistic perspective. This paper addresses this by comprehensively considering the natural meteorological and hydrological conditions of the watershed and the water management requirements of the watershed management agency when determining the drought threshold, while also taking into account the natural and social dimensions of drought. The identified drought events are more aligned with the current definition of drought. Simultaneously focusing on key indicators such as the response time, probability of occurrence, and trigger threshold of meteorological drought triggering hydrological drought, the paper incorporates covariates to consider runoff inconsistency caused by climate change and human activities and its impact on meteorological drought triggering hydrological drought. This approach can be used to assess hydrological drought in watersheds under different changing environments, with broad applicability. Considering that a single characteristic variable cannot fully characterize the features of meteorological drought triggering hydrological drought, a joint conditional probability distribution is constructed between the time series of hydrological drought intensity and meteorological drought characteristic variables (drought duration, drought intensity, and drought peak). This simultaneously considers multiple characteristic variables of drought events and their interrelationships, providing a more comprehensive description of meteorological drought. The dynamic relationship of hydrological drought propagation is studied to characterize the dynamic risk of meteorological drought triggering hydrological drought in a watershed from a mechanistic perspective, so as to truly reflect the actual drought risk situation and achieve the goal of natural drought propagation. Estimating the probability of occurrence of hydrological drought in a watershed using meteorological drought characteristic variables from a social coupling perspective.

[0007] Firstly, the present invention provides a natural A socially coupled watershed hydrological drought assessment method, characterized by the following steps: Collect precipitation and runoff observation data for the target watershed, and then use the collected precipitation and runoff observation data at control sections to analyze the precipitation data. Diagnosis of variations in runoff relationships; The drought index is calculated, which includes the meteorological drought index and the hydrological drought index that takes into account the impact of reservoir operation. Computation based on nature The drought threshold of social coupling, based on nature Socially coupled drought thresholds include standardized meteorological drought thresholds and standardized hydrological drought thresholds; Based on control section precipitation The results of runoff relationship variation diagnosis, combined with the drought index, are used to calculate the response time of hydrological drought to meteorological drought, based on natural... A socially coupled drought threshold is proposed, which estimates the conditional probability and propagation threshold of hydrological drought based on the Copula function, and assesses the risk of drought in the watershed based on the conditional probability and propagation threshold of hydrological drought.

[0008] According to the natural A socially coupled watershed hydrological drought assessment method, wherein the methods for collecting precipitation and runoff observation data of the target watershed include: Runoff observation data from hydrological stations in the target watershed were collected to obtain the monthly runoff sequence of the control section; precipitation observation data from meteorological stations in the target watershed were collected, and interpolation techniques were used to generate the corresponding monthly precipitation sequence for the runoff area of ​​the control section.

[0009] According to the natural A socially coupled watershed hydrological drought assessment method, wherein the collected precipitation and runoff observation data of the target watershed are used to control section precipitation. Methods for diagnosing runoff variation and identifying abrupt change points include: By selecting the sliding step size and sliding window length of the monthly precipitation and runoff series, the Copula function is used to calculate the precipitation for each sliding window sequentially. Runoff correlation coefficient; The optimal Copula function for each sliding window is determined sequentially, and the precipitation is obtained as the sliding step size increases. Runoff correlation coefficient sequence; The obtained precipitation Change point diagnosis was performed on the runoff correlation coefficient sequence to obtain abrupt change points.

[0010] According to the natural A socially coupled watershed hydrological drought assessment method, wherein the obtained precipitation Methods for identifying abrupt change points using runoff correlation coefficient sequences include: Construct a rank-based cumulative sum statistic to quantify precipitation. Two precipitation periods before and after any candidate point in the runoff correlation coefficient sequence The overall difference in runoff correlation coefficients yields the absolute values ​​of the statistics corresponding to all change points; Calculate the probability of the maximum absolute value of the change point statistic occurring, and compare the probability of the maximum absolute value of the change point statistic occurring with a preset significance level threshold: if the probability of the maximum absolute value of the change point statistic occurring is less than or equal to the preset significance level threshold, then the time corresponding to the maximum absolute value of the change point statistic is determined to be a change point.

[0011] According to the natural A socially coupled watershed hydrological drought assessment method, wherein the method for calculating the drought index includes: The cumulative probability distribution of monthly precipitation is obtained by fitting the monthly precipitation sequence; The standardized precipitation index (SPI), or meteorological drought index, is calculated by approximate algorithm after normalizing the monthly cumulative probability distribution. The reservoir index was used as a covariate, and the cumulative probability distribution of monthly runoff at the control section was calculated using the GAMLSS model. The cumulative probability distribution of monthly runoff was then normalized to obtain the non-uniform standardized runoff index NSSI, which is the hydrological drought index that takes into account the impact of reservoir scheduling.

[0012] According to the natural A socially coupled watershed hydrological drought assessment method, which calculates the hydrological drought index considering the impact of reservoir operation, includes the following sub-steps: Determine the non-consistent alternative combinations of runoff probability distribution models and covariates. The probability distribution parameters include location parameters, scale parameters, and shape parameters. At least one of the location parameters and scale parameters contains a covariate, which includes the reservoir index. By inputting the runoff probability distribution model and the non-consistent alternative combinations of covariates, as well as the cumulative runoff sequences for each month, into the GAMLSS model, the optimal non-consistent combination of the runoff probability distribution model and covariates is determined, and the monthly cumulative runoff probability distribution at the control section is obtained. The non-uniform standardized runoff index (NSSI) is obtained by normalizing the cumulative probability distribution of monthly runoff at the control section, which is a hydrological drought index that takes into account the impact of reservoir scheduling.

[0013] According to the natural A socially coupled watershed hydrological drought assessment method, the calculation of which is based on natural... Methods for determining drought thresholds based on social coupling include: The first preset percentile of the cumulative probability distribution of monthly precipitation is used as the meteorological drought threshold, and the meteorological drought threshold is transformed into a standardized meteorological drought threshold. The flow rate corresponding to the second preset percentile in the cumulative probability distribution of monthly runoff is used as the threshold for hydrological and drought natural aspects. The minimum discharge flow at the control section is used as the threshold for hydrological, drought, and social aspects. The outer bounds of the natural and social aspects of hydrological drought thresholds are used as the hydrological drought threshold. After normalization transformation of the hydrological drought threshold, the standardized hydrological drought threshold is calculated using an approximate algorithm.

[0014] According to the natural A socially coupled watershed hydrological drought assessment method, based on control section precipitation. Methods for calculating the response time of hydrological drought to meteorological drought, based on runoff relationship variation diagnostic results and combined with the drought index, include: For undiagnosed precipitation For the control sections where runoff relationships change abruptly, causal analysis is performed using the entire sequences of meteorological drought index and hydrological drought index at the control sections to determine the response time of hydrological drought to meteorological drought. For diagnosing precipitation The control section of the runoff relationship abrupt change point is used as the boundary to divide the meteorological drought index and hydrological drought index into two time periods before and after the change point for causal relationship analysis, and to determine the response time of hydrological drought to meteorological drought before and after the change point.

[0015] According to the natural A socially coupled watershed hydrological drought assessment method, wherein the method for conducting causal relationship analysis includes: The correlation coefficient between the meteorological drought index and the hydrological drought index was calculated using a convergent cross mapping (CCM) algorithm with time lag, and plotted on the time lag. In the correlation coefficient plane coordinate system, the x-coordinate corresponding to the highest point in the obtained scatter plot is taken as the response time of hydrological drought to meteorological drought.

[0016] According to the natural A socially coupled watershed hydrological drought assessment method, based on natural factors... Socially coupled drought thresholds and methods for estimating the conditional probability of hydrological drought occurrence based on Copula functions include: Based on nature The socially coupled drought threshold was used as a cutoff threshold to extract meteorological drought events and hydrological drought events from the drought index sequence, respectively. Considering the response time of hydrological drought to meteorological drought, the extracted meteorological drought events and hydrological drought events are matched to obtain meteorological drought data. For hydrological drought event pairs, extract the corresponding drought characteristic variables for each drought event pair; The drought characteristic variables obtained by matching are regarded as random variables. The marginal probability distribution parameters are fitted, and the most suitable marginal probability distribution is obtained according to the runoff probability distribution model and the optimal combination of covariates with non-consistent covariates. The Vine Copula function is used to process the multivariate joint distribution of drought characteristic variables. By decomposing them into paired Copula structures, the drought characteristic variables include meteorological drought duration, meteorological drought intensity, meteorological drought peak, and hydrological drought intensity. Based on the cumulative probability of the matched meteorological drought intensity, hydrological drought events are divided into four levels: mild, moderate, severe, and extreme. When the meteorological drought characteristic variables exceed the preset threshold, the conditional probability of triggering severe and extreme hydrological drought events is estimated in each sub-interval within the specified threshold range.

[0017] Secondly, the present invention also provides a natural A socially coupled watershed hydrological drought assessment system is used to implement the natural [measures] provided by the first aspect of this invention and any of its optional embodiments. Socially coupled watershed hydrological drought assessment methods include: precipitation The runoff relationship variation diagnosis module is used to collect precipitation and runoff observation data of the target watershed, and to control the precipitation at the cross-sections of the collected precipitation and runoff observation data of the target watershed. Diagnosis of variations in runoff relationships; The drought index calculation module is used to calculate the drought index, which includes the meteorological drought index and the hydrological drought index that takes into account the impact of reservoir operation. Based on nature A socially coupled drought threshold calculation module is used to calculate drought thresholds based on natural factors. The drought threshold of social coupling, based on nature Socially coupled drought thresholds include standardized meteorological drought thresholds and standardized hydrological drought thresholds; The watershed drought risk assessment module is used to assess precipitation based on control sections. The results of runoff relationship variation diagnosis, combined with the drought index, are used to calculate the response time of hydrological drought to meteorological drought, based on natural... A socially coupled drought threshold is proposed, which estimates the conditional probability and propagation threshold of hydrological drought based on the Copula function, and assesses the risk of drought in the watershed based on the conditional probability and propagation threshold of hydrological drought.

[0018] Thirdly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the natural functions described above. A socially coupled watershed hydrological drought assessment method.

[0019] Fourthly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the natural state as described in any of the above embodiments. A socially coupled watershed hydrological drought assessment method.

[0020] The present invention also provides a computer program product, comprising a computer program that, when executed by a processor, implements the natural language described above. A socially coupled watershed hydrological drought assessment method.

[0021] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention uses a sliding copula to control the precipitation at the control section. Runoff relationships are used to diagnose change points. For cross sections where change points are diagnosed, the response time of hydrological drought to meteorological drought is determined segmentally before and after the change point, and then matched with meteorological data. The marginal distribution of hydrological drought events and fitting drought characteristic variables is used to consider the impact of runoff inconsistency caused by human activities such as reservoir operation on the propagation characteristics of meteorological to hydrological drought and the parameters of Copula correlation function, so as to more accurately characterize the dynamic risk of meteorological drought triggering hydrological drought in the watershed. (2) The method for determining drought threshold in this invention comprehensively considers the meteorological and hydrological natural conditions of the basin and the water management requirements of the basin management agency, while taking into account the natural attributes and social dimensions of drought. The identified drought events are more in line with the current definition of drought. (3) This invention simultaneously focuses on key indicators such as response time, occurrence probability and trigger threshold of meteorological drought triggering hydrological drought. In the calculation process, by introducing covariates, it considers the runoff inconsistency caused by climate change and human activities and its impact on meteorological drought triggering hydrological drought. It can be used to assess watershed hydrological drought under changing conditions and has a wide range of applications. (4) Considering that a single feature variable cannot fully characterize the features of meteorological drought triggering hydrological drought, a joint conditional probability distribution between the time series of hydrological drought intensity and meteorological drought feature variables (drought duration, drought intensity, and drought peak) is constructed. At the same time, multiple feature variables of drought events and their interrelationships are considered to more comprehensively describe the meteorological drought. The dynamic relationship of hydrological drought propagation is used to characterize the dynamic risk of meteorological drought triggering hydrological drought in the basin from a mechanistic perspective, so as to truly reflect the actual drought risk situation. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in this 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 some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0023] Figure 1 This invention provides a natural A flowchart illustrating a socially coupled watershed hydrological drought assessment method; Figure 2 It is a natural method provided by the present invention. A socially coupled watershed hydrological drought assessment method using a sliding step size ξ Calculate precipitation for each sliding window sequentially. A schematic diagram of the Copula function process of a runoff sequence; Figure 3 It is a natural method provided by the present invention. Copula correlation coefficients of precipitation and runoff sequences at the Huangjiagang section under different sliding windows were calculated using a socially coupled watershed hydrological drought assessment method. Figure 4 It is a natural method provided by the present invention. A schematic diagram of the diagnostic results of the correlation coefficient change point of precipitation and runoff at Huangjiagang and Huangzhuang sections calculated by the socially coupled watershed hydrological drought assessment method; Figure 5 It is a natural method provided by the present invention. Schematic diagram of the SPI and NSSI sequences for (a) Huangjiagang and (b) Huangzhuang sections from 1957 to 2019, calculated using a socially coupled watershed hydrological drought assessment method; Figure 6 This invention provides a natural The computational natural resources involved in socially coupled watershed hydrological drought assessment methods A schematic diagram of the hydrological drought threshold of social coupling; Figure 7It is a natural method provided by the present invention. A schematic diagram illustrating the probability of extreme hydrological drought occurrence at Huangjiagang and Huangzhuang sections under different meteorological drought characteristics, obtained by a socially coupled watershed hydrological drought assessment method. Figure 8 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0024] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and should not be construed as limiting the scope of the invention.

[0025] In the description of this invention, it should be noted that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0026] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "connected" and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0027] like Figure 1 As shown, an embodiment of the present invention provides a natural A socially coupled watershed hydrological drought assessment method includes the following steps: S1, collect precipitation and runoff observation data of the target watershed, and control the precipitation at the cross-sections based on the collected precipitation and runoff observation data of the target watershed. Diagnosis of variations in runoff relationships; S2, calculate the drought index, which includes the meteorological drought index and the hydrological drought index considering the impact of reservoir operation; S3, calculation based on nature The drought threshold of social coupling, based on nature Socially coupled drought thresholds include standardized meteorological drought thresholds and standardized hydrological drought thresholds; S4, based on control section precipitation The results of runoff relationship variation diagnosis, combined with the drought index, are used to calculate the response time of hydrological drought to meteorological drought, based on natural... A socially coupled drought threshold is proposed, which estimates the conditional probability and propagation threshold of hydrological drought based on the Copula function, and assesses the risk of drought in the watershed based on the conditional probability and propagation threshold of hydrological drought.

[0028] This invention comprehensively considers the meteorological and hydrological conditions of the watershed and the water management requirements of the watershed management agency when determining the drought threshold, while also taking into account the natural attributes and social dimensions of drought. It simultaneously focuses on key indicators such as the response time, probability of occurrence, and trigger threshold of meteorological drought triggering hydrological drought. In the calculation process, it introduces covariates to consider the runoff inconsistency caused by climate change and human activities and their impact on meteorological drought triggering hydrological drought. By constructing a joint conditional probability distribution between the time series of hydrological drought intensity and meteorological drought characteristic variables (drought duration, drought intensity, and drought peak), and considering multiple characteristic variables of drought events and their interrelationships, it provides a more comprehensive description of meteorological drought. The dynamic relationship of hydrological drought propagation is studied to address the current watershed hydrological drought assessment methods, which tend to focus solely on the natural or social dimensions of drought in setting drought thresholds, neglecting the role of drought as a natural phenomenon. The dual nature of socially coupled disasters, coupled with a lack of consideration for the impact of runoff inconsistencies caused by human activities such as reservoir management on the propagation characteristics of meteorological to hydrological drought and the parameters of Copula-related functions, fails to characterize the dynamic risk of meteorological drought triggering hydrological drought in watersheds from a mechanistic perspective. This approach can make identified drought events more aligned with the current definition of drought, can be used for assessing hydrological drought in watersheds under changing environments, has a wide range of applications, and can more comprehensively describe meteorological drought. The dynamic relationship of hydrological drought propagation is studied to characterize the dynamic risk of meteorological drought triggering hydrological drought in a watershed from a mechanistic perspective, so as to truly reflect the actual drought risk situation and achieve the goal of natural drought propagation. Estimating the probability of occurrence of hydrological drought in a watershed using meteorological drought characteristic variables from a social coupling perspective.

[0029] In one specific implementation, the Huangjiagang and Huangzhuang sections of the main stream of the Hanjiang River, a first-level tributary of the Yangtze River, are used as examples of a natural [method / information] provided in this embodiment of the invention. An example is given to illustrate the socially coupled watershed hydrological drought assessment method.

[0030] In step S1 of this embodiment of the invention, precipitation observation data and runoff observation data of the target watershed are collected, and the collected precipitation observation data and runoff observation data of the target watershed are used to control the precipitation at the cross-section. Diagnostic analysis of runoff relationship variations reveals abrupt change points.

[0031] Considering the impact of reservoir regulation on runoff, the correlation between watershed precipitation and runoff varies over time, and the Copula function correlation coefficient used to describe the relationship between precipitation and runoff may also be time-varying. This is to diagnose watershed precipitation... To address the inconsistency in runoff-precipitation relationships, a Copula function is used to construct the correlation structure between precipitation and runoff. When the correlation coefficient exhibits a significant jump, it is determined that the precipitation at that cross-section... The runoff relationship has changed; the correlation coefficient includes the significance level.

[0032] In step S1 of this embodiment of the invention, the method for collecting precipitation observation data and runoff observation data of the target watershed includes: Runoff observation data from hydrological stations in the target watershed were collected to obtain the monthly runoff sequence of the control section; precipitation observation data from meteorological stations in the target watershed were collected, and interpolation techniques were used to generate the corresponding monthly precipitation sequence for the runoff area of ​​the control section. Continuing from the previous example, the Huangjiagang section is located downstream of the Danjiangkou Reservoir on the main stream of the Han River. It houses the Huangjiagang National Basic Hydrological Station and controls a runoff area of ​​95,200 km². 2 The Huangzhuang section marks the boundary between the middle and lower reaches of the Han River. The Huangzhuang National Basic Hydrological Station is located there, controlling a runoff area of ​​142,100 km². 2 Monthly runoff observation data from 1957 to 2019 were collected from two hydrological stations, Huangjiagang and Huangzhuang; monthly precipitation observation data from 1957 to 2019 were collected from 22 meteorological stations in the runoff area controlled by the Huangzhuang section and its surrounding areas, including Taibai, Lueyang, Liuba, Hanzhong, Foping, Ningqiang, Shiquan, Wanyuan, Zhenba, Zhen'an, Fengjie, Badong, Shangnan, Xixia, Yunxi, Fangxian, Xingshan, Nanyang, Zaoyang, Tongbai, Laohekou, and Xiangyang. Kriging interpolation was used to generate a 5km×5km resolution grid precipitation sequence for the runoff area controlled by the section. The areal average precipitation sequence, i.e., the monthly precipitation sequence, for the runoff area controlled by the Huangjiagang and Huangzhuang sections was calculated.

[0033] No. k Monthly precipitation sequence of the sub-basin corresponding to the control section catchment area P k ={ p k1 , p k2 , p k3 , …, p kN}, N Given the length of the precipitation sequence, the corresponding monthly runoff sequence is: Q k ={ q k1 , q k2 , qk3 , …, q kN}

[0034] In step S1 of this embodiment of the invention, the precipitation observation data and runoff observation data of the collected target watershed are controlled at the section precipitation. Methods for diagnosing runoff variation and identifying abrupt change points include: S11, Select the sliding step size and sliding window length for the monthly precipitation and runoff sequences, and use the Copula function to calculate the precipitation for each sliding window sequentially. Runoff correlation coefficient; Typically, a smaller window is selected for initial diagnosis, and then the window length is gradually increased for precise change point detection, avoiding omissions caused by an inappropriate sliding window length. This is because precipitation needs to be monitored. Years in which runoff variation occurs, sliding step size ξ Select December, slide window ω Set the step size to an integer multiple of the step size, such as 1 (12 months), 2 (24 months), 5 (60 months), etc. The sliding window length for both series should remain consistent and constant throughout the sliding process; start from the first data point of the time series, such as... Figure 2 As shown, with sliding step size ξ The time series is sequentially slid forward through the windows until the end, and the precipitation in each sliding window is calculated using the Copula function. Runoff correlation coefficients, Copula functions include the Gumbel Copula function, the Frank Copula function, and the Clayton Copula function, where the first... n The joint probability distribution of monthly precipitation and runoff sequences for each window, i.e., the precipitation in the nth window. Correlation coefficient of monthly runoff: (1) in, Monthly runoff sequence Q k In the n The edge distribution function of each window Monthly precipitation sequence P k In the n The edge distribution function of each window and marginal distribution and The distribution parameters, and They represent Q k and Pk In the n Edge probability distribution of each window C (·) represents the Copula function. For the first n The parameters of the Copula function for each window.

[0035] Continuing from the previous example, we select the sliding step size for the monthly precipitation series and the monthly runoff series. ξ Both are 12 months, sliding window length ω The sliding step sizes are 1, 2, 3, 5, 10, and 20 times the actual sliding step size, starting from the first data point in the time series, with each sliding step size... ξ Move window backward ω Precipitation under different sliding window lengths is calculated using equation (1) until the end of the time series. Runoff correlation coefficient. Taking Huangjiagang Station as an example, such as... Figure 3 As shown, precipitation is calculated under sliding windows of different lengths. Runoff correlation coefficient, when the sliding window length ω When the precipitation period is 36 months, the Huangjiagang station recorded precipitation from 1957 to 2019. The variation points in runoff relationship are obvious, and the selected sliding window length is 36 months (i.e., 3 times the sliding step length).

[0036] S12, determine the optimal Copula function for each sliding window in sequence, and obtain the precipitation as it slides with the sliding step size. Runoff correlation coefficient sequence; Using SBC (Schwarz information criteria, Bayesian information criterion) and Probability Integral Transform (PIT) as evaluation indicators for fitting performance, the optimal Copula function for each window is determined sequentially from the candidate Archimedes Copula functions (Gumbel Copula, Frank Copula, and Clayton Copula). The Copula function with the smallest SBC is selected as the optimal Copula function for that window, and the correlation coefficient sequence sliding with the step size is obtained.

[0037] Continuing from the previous example, based on the fitting results of the three alternative Archimedes Copula functions to the correlation structure of precipitation and runoff at the Huangjiagang section, the results show that the Gumbel Copula function has the best fitting effect. The PIT test results show that the selected Copula function has a good fit to the correlation structure of precipitation and runoff, thus obtaining the correlation coefficient sequence for each window.

[0038] S13, using the change point test method to analyze the obtained precipitation Change point diagnosis was performed on the runoff correlation coefficient sequence to obtain abrupt change points.

[0039] In step S1 of this embodiment of the invention, the change point test method is used to analyze the obtained precipitation. Methods for identifying abrupt change points using runoff correlation coefficient sequences include: Precipitation is quantified by constructing a rank-based cumulative sum statistic. Two precipitation periods before and after any candidate point in the runoff correlation coefficient sequence The overall difference in runoff correlation coefficients is used to obtain the absolute values ​​of the statistics corresponding to all change points, and the time corresponding to the change point with the largest absolute value of the statistics is determined as the change point where the sequence is most likely to undergo a sudden change. After determining the most likely location of the change point where a sudden change is to occur, it is necessary to further evaluate the statistical significance of the maximum absolute value of the change point statistic. This involves calculating the probability of the maximum absolute value of the change point statistic occurring (i.e., the probability of its occurrence). p The value is compared with a preset significance level threshold (e.g., 0.05): the probability of the absolute value of the change point statistic reaching its maximum is calculated. p If the value is less than or equal to the preset significance level threshold, the null hypothesis that "the sequence has no change point" is rejected, and the change point is determined to be statistically significant. The time corresponding to the time with the largest absolute value of the change point statistic is determined to be the change point.

[0040] In the embodiments of the present invention, the first is used to verify the... k Control section variation points t Taking the corresponding significance level as an example, to control the precipitation at the cross-section Diagnosis of variations in runoff relationships.

[0041] Pettitt's change-point test uses the following formula to test the first... k Control section variation points t Corresponding statistics and significance level : (2) (3) Let the significance level threshold be 0.05. n It is the sequence length, when the significance level is... At that time, it was assumed that the Copula parameter sequence was at the turning point. t There is no significant mutation point at this location; when When, the Copula parameter sequence at the change point t There is a significant mutation point at point 1, indicating that the first... k Cross-sectional precipitation Runoff relationship at a turning point t The mutation occurred at that location.

[0042] Continuing from the previous example, such as Figure 4 As shown, the Petitt change point test was used to diagnose the change points in the Copula correlation coefficient sequences of precipitation and runoff series at Huangjiagang and Huangzhuang sections. The year of change was found to be 1975. Literature review indicates that the preliminary works of the Danjiangkou Reservoir were completed in 1974, and the following year (1975), the precipitation and runoff relationships at Huangjiagang and Huangzhuang sections experienced abrupt changes. This suggests that the construction of the Danjiangkou Reservoir was the cause of the abrupt changes in precipitation at these two sections. The main reasons for abrupt changes in runoff relationship.

[0043] In step S2 of this embodiment of the invention, a drought index is calculated, which includes a meteorological drought index and a hydrological drought index that takes into account the impact of reservoir scheduling.

[0044] In step S2 of this embodiment of the invention, the method for calculating the meteorological drought index in the drought index includes: Monthly precipitation sequence P k The cumulative probability distribution of monthly precipitation was obtained by fitting. ; After normalizing the monthly cumulative probability distribution of precipitation, the Standardized Precipitation Index (SPI) is calculated using the Abramowitz-Stegun approximation algorithm. SPI is essentially a meteorological drought index. (4) (5) in, p It's precipitation. Z It is a variable used to calculate the standardized precipitation index. z It is the default value. G (·) represents the function symbol. z = p Indicates the observed precipitation; δ As an intermediate variable; a 0, a 1, a 2, b 1, b 2, b 3 is a constant; when When ≤ 0.50, κ = 1. When When >0.50, the value in equation (5) will be... Replace with ,and κ =1; for Z = zThe cumulative probability distribution.

[0045] Following the previous example, based on the average monthly precipitation sequence of the Huangjiagang section control runoff area from 1957 to 2019, the cumulative probability distribution of monthly precipitation in Huangjiagang and Huangzhuang sections was fitted, and the corresponding meteorological drought index (SPI) sequence was calculated by equations (2) and (3).

[0046] In step S2 of this embodiment of the invention, the method for calculating the hydrological drought index considering the impact of reservoir scheduling in the drought index includes: using the reservoir index as a covariate, using the GAMLSS model to calculate the monthly runoff cumulative probability distribution of the control section, and performing normalization processing on the monthly runoff cumulative probability distribution to obtain the non-uniform standardized runoff index NSSI, that is, the hydrological drought index considering the impact of reservoir scheduling.

[0047] Due to the impact of reservoirs on runoff conditions, the consistency assumptions relied upon by the traditional SSI (Standardized Streamflow Index) are no longer applicable. The Non-Stationary Standardized Streamflow Index (NSSI) is selected to characterize the hydrological drought index. In this embodiment of the invention, the GAMLSS (Generalized Additive Models for Location, Scale and Shape) model is used to consider runoff inconsistency in calculating the hydrological drought index.

[0048] In step S2 of this embodiment of the invention, the method for calculating the hydrological drought index that takes into account the impact of reservoir scheduling includes the following sub-steps: S21. Determine the non-consistent alternative combinations of the runoff probability distribution model and covariates. The probability distribution parameters include location parameters, scale parameters, and shape parameters. At least one of the location parameters and scale parameters contains a covariate, which includes the reservoir index.

[0049] The runoff probability distribution models in the alternative combinations are six commonly used probability distribution models: two-parameter probability distribution models Gamma (GA), Normal (NO), Gumbel (GU), Logistic (LO), and Weibull (WEI), and a three-parameter probability distribution model Generalized Extreme Value (GEV). The probability distribution parameters include location parameters. μ Scale parameter σ and shape parameter ζAs shown in Table 1, there are 6 consistent alternative combinations, meaning all probability distribution parameters are fixed and contain no covariates; there are 18 inconsistent alternative combinations, where at least one probability distribution parameter of the location or scale parameter contains a covariate, while inconsistency is not considered for the shape parameter. Covariates include temperature and precipitation considering the impact of climate change, and reservoir index and water consumption considering the impact of human activities. Taking the reservoir index as an example, the calculation formula is as follows: (6) Among them, RI k for k The reservoir index at the cross-section; A i for i The runoff area controlled by the reservoir; V i for i Total reservoir capacity; A k for k The runoff area controlled by cross-section; W k for k Multi-year average runoff of the cross section; N k for k The number of associated reservoirs in a given year at a cross-section.

[0050] Table 1. Alternative combinations of runoff probability distributions selected in this invention.

[0051] Note: Z i = ( z 1, z 2,…, z i ) represents a vector of covariates.

[0052] S22. Input the runoff probability distribution model and the non-consistent alternative combination schemes of covariates, as well as the cumulative runoff sequences of each month, into the GAMLSS model to determine the optimal non-consistent combination scheme of the runoff probability distribution model and covariates, and obtain the monthly cumulative runoff probability distribution of the control section.

[0053] The probability distribution parameters of the alternative combinations are estimated using the maximum likelihood estimation method. Based on the SBC criterion, the candidate combination scheme with the minimum SBC value for each station was selected as the optimal combination scheme; Worm plots and normal QQ plots were used for graphical diagnosis, and the KS test was used to evaluate the model fit, with the following statistic: (7) In the formula, For the accumulated probability of experience, from the formula t / ( n +1) is calculated to obtain, n It is the length of the monthly runoff sequence. t It is a serial number; G (t) This represents the theoretical cumulative probability in ascending order. When the KS test statistic is less than the critical value (e.g., 0.05), it indicates that the measured samples conform to the selected model, meaning that the selected model has passed the goodness-of-fit test.

[0054] Continuing the previous example, the reservoir index is used as a covariate to consider the impact of upstream reservoir regulation. The GAMLSS model is used to calculate the monthly cumulative runoff probability distributions at the Huangjiagang and Huangzhuang sections. Based on the SBC criterion, the optimal probability distribution functions for the Huangjiagang section from January to December are GEV, NO, GA, GA, GEV, GA, GA, GA, GA, GA, GA, GA, GEV, and for the Huangzhuang section, they are NO, LO, NO, LO, GEV, GEV, GEV, GEV, GEV, LO, NO, NO. Taking January and July as examples, a graphical diagnostic is performed using Worm plots and normal QQ plots. Most points in the residual normal QQ plot are distributed near the 1:1 line, and all points in the Worm plot are distributed within the 95% confidence interval, indicating that the selected model has a good fit. The KS test is used to evaluate the model's goodness of fit. The KS test results for each month are... p The values ​​are all greater than 0.05, indicating that the selected model has a good fit.

[0055] S23, the non-uniform standardized runoff index NSSI is obtained by normalizing the cumulative probability distribution of monthly runoff at the control section, which is the hydrological drought index that takes into account the impact of reservoir scheduling.

[0056] The cumulative probability distribution of runoff estimated by the GAMLSS model for each month is normalized, and the cumulative probability distribution in equation (5) is replaced with... Substituting into equation (4), the hydrological drought index NSSI is calculated, that is: (8) in, For the first k Station No. l Monthly Given The cumulative probability distribution of runoff.

[0057] Following the previous example, the 1957 values ​​for the Huangjiagang and Huangzhuang sections were calculated using equation (8). The monthly drought index NSSI for 2019, along with SPI, is plotted and displayed on [platform name missing]. Figure 5 middle.

[0058] In step S3 of this embodiment of the invention, the calculation is based on natural The drought threshold of social coupling, based on nature Socially coupled drought thresholds include standardized meteorological drought thresholds and standardized hydrological drought thresholds.

[0059] Drought has both natural and social attributes; its evolution into a severe drought disaster depends on its impact on local people's livelihoods and the environment. Existing research largely focuses on the natural perspective, employing fixed thresholds (such as...). 0.5) serves as the cutoff threshold; a drought event is defined as any time when the drought index (e.g., NSSI) value falls below this threshold. Considering the runoff regulation within water supply systems, drought assessment should not solely focus on the magnitude of runoff at a given time period, but rather on situations where runoff is less than the actual water demand. In recent years, my country has formulated and issued water allocation plans for major river basins, taking into account ecological and downstream water demand for domestic, industrial, irrigation, and navigation purposes, and determining the minimum discharge flow at control sections. This indicates a greater emphasis on the social dimension of drought. The Threshold Level Method considers the water allocation needs of river basin management agencies from a social perspective, but this method neglects local hydrological and climatic conditions. The following proposes a natural... A method for determining drought thresholds from the perspective of social coupling.

[0060] In step S3 of this embodiment of the invention, the calculation is based on natural The method for standardizing meteorological drought thresholds in socially coupled drought thresholds includes: using the first preset percentile of the monthly cumulative precipitation probability distribution as the meteorological drought threshold, and transforming the meteorological drought threshold into a standardized meteorological drought threshold.

[0061] The preset percentile of the obtained monthly precipitation cumulative probability distribution is used as the meteorological drought threshold, and it is converted into a standardized threshold by formula (2).

[0062] (9) in, It is a standardized meteorological drought threshold used to extract meteorological drought events from SPI sequences.

[0063] Continuing from the previous example, the precipitation corresponding to the 50th quantile of the cumulative monthly precipitation probability distribution at the Huangjiagang and Huangzhuang sections is taken as the meteorological drought threshold. The corresponding standardized drought threshold is 0, meaning the meteorological drought threshold is fixed as a constant.

[0064] In step S3 of this embodiment of the invention, the calculation is based on natural Methods for standardizing hydrological drought thresholds in socially coupled drought thresholds include: S31, the flow rate corresponding to the second preset percentile in the cumulative probability distribution of monthly runoff is used as the threshold for hydrological drought natural aspects; S32, the minimum discharge flow at the control section is used as the threshold for hydrological, drought and social aspects; S33 uses the outer envelope of the natural and social aspects of hydrological drought thresholds as the hydrological drought threshold. After normalizing the hydrological drought threshold, an approximate algorithm is used to calculate the standardized hydrological drought threshold.

[0065] Generally, below-normal flow rates indicate a drought event at the cross-section from a natural perspective. The flow rate corresponding to the Φth percentile of the cumulative monthly runoff distribution calculated in step 2 is used as the natural aspect threshold for hydrological drought, reflecting local hydrological and meteorological conditions. Since flow rates below the minimum discharge rate determined by the watershed management agency will lead to regional water supply and demand imbalances, reflecting the occurrence of drought events from a social perspective, the minimum discharge rate at the cross-section is used as the social aspect threshold for hydrological drought.

[0066] like Figure 6 As shown, the outer bounds of the natural and social aspects of hydrological drought thresholds are used as the natural... The socially coupled hydrological drought threshold can comprehensively reflect the interaction between the natural attributes of drought events and human activities, that is, whether water resource management can provide sufficient water supply to meet water demand. In nature... In the socially coupled hydrological drought threshold, the hydrological drought threshold is determined separately for each month (i.e., there are 12 thresholds per year): (10) in, yes k The flow rate value corresponding to the percentile of section Φ. This is the minimum discharge flow at the cross-section determined by the river basin management agency. Therefore, in natural... The standardized hydrological drought threshold from the perspective of social coupling can be obtained by conversion using equation (4): (11) in, It is a standardized hydrological drought threshold used to extract hydrological drought events from NSSI sequences.

[0067] Following the previous example, according to the "Hanjiang River Basin Water Dispatch Plan (Trial)" issued by the Yangtze River Water Resources Commission of the Ministry of Water Resources in 2020, considering the ecological and downstream water demand for domestic, industrial, irrigation, and navigation purposes, the minimum discharge flow at the Huangjiagang section is determined to be 490 m³ / s. 3 / s, the minimum discharge flow at the Huangzhuang section is 500 m³ / s.3 / s, respectively, are used as the hydrological drought thresholds for these two sections from a social perspective. The flow corresponding to the 50th quantile of the cumulative precipitation distribution function for each month at Huangjiagang and Huangzhuang sections is used as the hydrological drought threshold from a natural perspective. The larger of the two thresholds for each month is taken according to Equation (10), and Equation (11) is used to convert it into a standardized hydrological drought threshold. The cumulative runoff probability distribution corresponding to the minimum discharge flow is different in different months, and the resulting hydrological drought threshold is different in each month, i.e., a variable threshold; compared with before the change point (1957–1974), after the change point (1975–2019), the runoff at Huangjiagang and Huangzhuang sections is affected by the regulation of the upstream Danjiangkou Reservoir, and the runoff changes in each month. The hydrological drought thresholds before and after the change point are also different, see Figure 5 The red line in the middle.

[0068] In step S4 of this embodiment of the invention, based on the control section precipitation The results of runoff relationship variation diagnosis, combined with the drought index, are used to calculate the response time of hydrological drought to meteorological drought, based on natural... A socially coupled drought threshold is proposed, which estimates the conditional probability and propagation threshold of hydrological drought based on the Copula function, and assesses the risk of drought in the watershed based on the conditional probability and propagation threshold of hydrological drought.

[0069] Meteorological drought and hydrological drought do not occur simultaneously; there is often a time lag between them. Matching meteorological drought... Before addressing a hydrological drought event, the triggering interval of the corresponding meteorological drought event should be determined. Considering that hydrological drought usually lags behind meteorological drought and is considered to have a time-delayed unidirectional causal relationship, a time-delayed CCM is used to determine the response time (on a monthly basis) of the hydrological drought index (NSSI calculated in step S22) to the meteorological drought index (SPI calculated in step S21), i.e., the response time of hydrological drought to meteorological drought.

[0070] In step S4 of this embodiment of the invention, the step based on the control section precipitation... Methods for calculating the response time of hydrological drought to meteorological drought, based on runoff relationship variation diagnostic results and combined with the drought index, include: For undiagnosed precipitation For cross sections where runoff relationships change abruptly, the entire sequences of meteorological drought index and hydrological drought index at that cross section are used to determine the response time of hydrological drought to meteorological drought through causal relationship analysis. For diagnosing precipitation The cross-section of the runoff relationship abrupt change point is used as the boundary to divide the meteorological drought index sequence and the hydrological drought index sequence into two time periods: before and after the change point. The response time of hydrological drought to meteorological drought before and after the change point is determined by causal relationship analysis.

[0071] In step S4 of this embodiment of the invention, the method for causal relationship analysis includes: calculating the correlation coefficient between the meteorological drought index series and the hydrological drought index series using CCM with lag, and plotting the correlation coefficient on the lag time. In the correlation coefficient plane coordinate system, the x-coordinate corresponding to the highest point in the obtained scatter plot is taken as the response time of hydrological drought to meteorological drought.

[0072] Using the diagnosed mutation point as the boundary, the SPI and NSSI sequences at the control section were divided into two time periods: before and after the mutation point, for causal relationship analysis. X and Y were used to simplify the SPI and NSSI time series variables, respectively, and lags representing the time lag characteristics of the causal relationship were added. The lagged coordinate vectors of variables X and Y are constructed as follows: (12) (13) in, E The embedding dimension is used to determine the optimal embedding dimension for manifold reconstruction through cross-validation. τ For reconstructing the manifold, the optimal sampling interval for a low-dimensional system is... τ It is usually 1 or 2.

[0073] The correlation coefficient between the meteorological drought index series and the hydrological drought index series was calculated using CCM with time lag, and plotted on the time lag. In the correlation coefficient plane coordinate system, the x-coordinate corresponding to the highest point in the scatter plot is taken as the response time of hydrological drought to meteorological drought.

[0074] The correlation coefficient between the two sequences (SPI and NSSI) was calculated using the CCM package and plotted on the time delay. In the correlation coefficient planar coordinate system, the x-axis of the highest point in the scatter plot represents the response time of hydrological drought (NSSI) to meteorological drought (SPI). T p .

[0075] Following the previous example, the obtained SPI and NSSI sequences were divided into pre-change (1957–1974) and post-change (1975–2019) periods based on the change point diagnosis results for causal relationship analysis. Sampling interval. τ Setting it to 1 indicates the optimal number of embedding dimensions. E Cross-validation was used to determine the location of the Huangjiagang section before and after the point change. E The values ​​are 4 and 5 respectively, before and after the Huangzhuang section change point. E The values ​​are 4 and 2 respectively. Different time lags before and after the change point at Royal Port and Huangzhuang sections. ( June The causal relationship (in June) indicates that the Huangjiagang section changed before... = The strongest causal relationship between meteorological drought and hydrological drought was observed at 3:00 AM, with a correlation coefficient of 0.46. Therefore, it was determined that the response time of hydrological drought to meteorological drought before the change point at the Huangjiagang section was 3 months. Similarly, based on the CCM results, the response time after the change point was determined to be 1 month. The response times of hydrological drought to meteorological drought before and after the change point at the Huangzhuang section were 2 months and 1 month, respectively.

[0076] The basis is based on nature Socially coupled drought thresholds and methods for estimating the conditional probability of hydrological drought occurrence based on Copula functions include: S41 will be based on nature The socially coupled drought threshold was used as a cutoff threshold to extract meteorological drought events and hydrological drought events from the drought index sequence.

[0077] The specific method of step S41 includes: using a standardized meteorological drought threshold as a cutoff threshold to extract meteorological drought events from the meteorological drought index sequence, and using a standardized hydrological drought threshold as a cutoff threshold to extract hydrological drought events from the hydrological drought index sequence.

[0078] Three characteristic variables for meteorological and hydrological drought events were extracted from the SPI / NSSI series using runs theory, and analyzed using pre-defined standardized thresholds. Drought duration, drought intensity, and drought peak are the three basic characteristics describing drought events. Drought duration is the length of time the SPI / NSSI index is below the pre-defined standardized drought threshold. Drought intensity and drought peak are defined by the cumulative and maximum SPI / NSSI deficit during the corresponding drought duration, respectively. Simultaneously, a merge-and-remove method was used to handle "associated drought" and "weak drought" events to avoid these two types of special drought events affecting the statistical properties of the drought characteristic variables.

[0079] Continuing with the previous example, using run theory, the meteorological drought threshold obtained in step S31 is used as the cutoff threshold to extract meteorological drought events and their characteristic variables (including drought duration, drought intensity, and drought peak) from the SPI sequence obtained in step S21. The hydrological drought threshold obtained in step S32 is used as the cutoff threshold to extract hydrological drought events and their characteristic variables from the NSSI sequence obtained in step S23.

[0080] S42, considering the response time of hydrological drought to meteorological drought, the extracted meteorological drought events and hydrological drought events are matched to obtain meteorological drought data. For hydrological drought event pairs, extract the corresponding drought characteristic variables for each drought event pair.

[0081] Hydrological drought events are a reflection of the past T p The response results to meteorological drought conditions within one month; for a certain hydrological drought event, the corresponding trigger interval is the period before the start of the hydrological drought event. T p The period from the end of the drought event to the end of the current event. If the triggering intervals of two consecutive meteorological drought events overlap, the triggering interval of the later drought event is the period from the end of the previous event to the end of the current event. Considering the impact of hydrological drought on the response time of meteorological drought, based on precipitation... Different scenarios are diagnosed by identifying runoff relationship change points. Extracted meteorological and hydrological drought events are matched to obtain meteorological-hydrological drought event pairs. Corresponding drought characteristics of these pairs are extracted, including drought duration, drought intensity, and peak drought value. For cases where precipitation is not diagnosed... The cross-sections at abrupt changes in runoff relationships are used to match meteorological drought events with hydrological drought events as a whole, yielding meteorological... Hydrological drought events are relevant to the diagnosis of precipitation. The cross-section at the point of abrupt change in runoff relationship is used as the boundary to divide the meteorological and hydrological drought event into two time periods: before and after the change point. Meteorological data for both periods are then matched. Hydrological drought events.

[0082] S43, treating the matched drought characteristic variables as random variables, fitting the marginal probability distribution parameters, and obtaining the most suitable marginal probability distribution based on the runoff probability distribution model and the optimal combination scheme of the non-consistent covariates.

[0083] For undiagnosed precipitation For cross-sections at points of abrupt changes in runoff relationships, the matched sequences of meteorological and hydrological drought characteristic variables are fitted as a whole to obtain the most suitable marginal probability distribution for each characteristic variable; this is used to diagnose precipitation... The cross-section of the runoff relationship abrupt change point is used as the boundary to divide meteorological drought events and hydrological drought events into two time periods: before and after the change point. The most suitable marginal probability distributions of drought characteristic variables before and after the change point are then fitted.

[0084] Based on the SBC criterion, the most suitable marginal probability distribution is selected from the candidate distribution functions chosen in step S2, and used to describe the duration of the meteorological drought. Z 1. Intensity Z 2 and peak Z 3 and hydrological drought intensity Z4. The parameters of each marginal probability distribution were estimated using the maximum likelihood method, and the goodness of fit of these marginal probability distributions was tested using the KS test. The marginal distribution statistical parameters of meteorological drought characteristic variables were set as constants, while the parameters of hydrological drought intensity were set as constants. Z The probability distribution parameters of 4 are considered to change over time, and their inconsistency is addressed by introducing relevant covariates. The optimal function is selected from three Archimedes Copula functions (Clayton, Gumbel, and Frank), and the marginal probability distributions are combined pairwise using a hierarchical Pair Copula decomposition strategy. The effectiveness is estimated using the PIT test.

[0085] S44 uses the Vine Copula function to process the multivariate joint distribution of drought characteristic variables. By decomposing it into a pairwise Copula structure, the drought characteristic variables include meteorological drought duration, meteorological drought intensity, meteorological drought peak, and hydrological drought intensity. Based on the cumulative probability of the matched meteorological drought intensity, the hydrological drought event is divided into four levels: mild, moderate, severe, and extreme. When the meteorological drought characteristic variable exceeds the preset threshold, the conditional probability of triggering severe and extreme hydrological drought events is estimated in each sub-interval within the specified threshold range.

[0086] The Vine Copula function is used to handle the multivariate joint distribution of drought characteristic variables. By decomposing it into paired Copula structures, the high-dimensional and complex joint distribution is decomposed into a product of a series of paired binary Copulas. This method overcomes the limitations of traditional multivariate Copulas in handling high-dimensional data, such as insufficient flexibility and difficulty in parameter estimation, and can more accurately characterize the nonlinearity and tail correlation between variables.

[0087] For a n dimensional vector v = ( v 1, v 2, ..., v n The conditional distribution function based on the regularized Vine Copula can be expressed as follows: (14) in, v n ( n =1, 2, ..., N () is any component of vector v. v - i Indicates the removal of the first i The vector v after each component, C z , vkIt is a bivariate pair copula function. Matching meteorological drought characteristic variables. Z 1. Z 2 and Z 3 serves as a condition to match the intensity of hydrological drought. Z 4 is used as the target. Typically, based on the cumulative probability of the matched meteorological drought intensity, hydrological drought is divided into four levels: mild, moderate, severe, and extreme, with corresponding probability intervals of [0, 0.50), [0.50, 0.75), [0.75, 0.90), and [0.90, 1.00], respectively. Furthermore, the minimum and maximum values ​​of each drought characteristic variable matched by MD are set as the lower and upper limits of the corresponding threshold range, with a step size of 0.10. For cases where the meteorological drought characteristic variable exceeds a given value, the conditional probability of triggering severe and extreme hydrological drought events is estimated within each sub-interval of the specified threshold range, i.e., when... Z 1≥ z 1. Z 2≥ z 2. Z 3≥ z At 3 o'clock, Z 4≥ z The conditional probability of 4 can be expressed as: (15) in, express The abbreviation of , where the symbol "^" indicates that the variable under that symbol changes with the covariate; for z s The marginal cumulative probability distribution, subscript s = 1, 2, 3, 4.

[0088] Continuing from the previous example, based on the response time of hydrological drought to meteorological drought determined in step S41, the extracted meteorological drought events and hydrological drought events are matched to obtain the meteorological data for the Huangjiagang and Huangzhuang sections, respectively. For hydrological drought event pairs, the matched sequences of meteorological drought and hydrological drought characteristic variables are obtained. The marginal probability distributions of each characteristic variable are fitted according to the SBC criterion, and the pairwise combinations are used to obtain the Pair Copula. The goodness of fit of the marginal probability distribution is tested by PIT. Under given meteorological drought conditions, the probability of occurrence of different levels of hydrological drought before and after the change point of Huangjiagang and Huangzhuang sections is estimated by Equation (15). Taking extreme drought as an example, the probability of occurrence of extreme hydrological drought at Huangjiagang and Huangzhuang sections under different meteorological drought characteristics is shown in the figure. Figure 7 .

[0089] The method for estimating the propagation threshold of hydrological drought based on the Copula function includes: Assuming a confidence level of 0.95 for meteorological drought triggering the hydrological drought threshold, the meteorological drought characteristic variable value corresponding to a conditional probability of 0.95 is considered the propagation threshold for triggering a hydrological drought event of that level. The risk of drought occurrence in the watershed is assessed based on the conditional probability and propagation threshold of hydrological drought. By comparing the conditional probability and triggering threshold of meteorological drought triggering hydrological drought before and after the change point, the rationality of the current water allocation plan is evaluated, guiding the formulation of watershed water allocation plans.

[0090] If, after the change point, the conditional probability of meteorological drought triggering hydrological drought decreases while the trigger threshold for hydrological drought increases, it indicates that the current water allocation plan is reasonable in mitigating the spread of drought. Conversely, if, after the change point, the conditional probability of meteorological drought triggering hydrological drought increases while the trigger threshold for hydrological drought decreases, it indicates that the current water allocation plan is unreasonable in mitigating the spread of drought and needs adjustment.

[0091] Continuing the previous example, when the conditional probability equals 0.95, the corresponding meteorological drought characteristic variable value is the trigger threshold for that level of hydrological drought event. Each hydrological drought trigger threshold includes three meteorological drought characteristic variable values: drought duration, drought intensity, and drought peak value. For example, for a meteorological drought event with a duration of 4 months, when the drought peak value... z When 3=2.5, due to the influence of the Danjiangkou Reservoir's water storage regulation, the drought intensity threshold changed from before the change point (1957). The value increased from 5.34 in 1974 to after the turning point (1975). The drought peak was 5.62 in 2019; while when the drought peak... z When 3 = 3.0, the threshold decreases from 4.44 before the change point to 4.11 after the change point. This indicates that the regulation of the Danjiangkou Reservoir affects the threshold for triggering hydrological drought in downstream sections.

[0092] The following describes a natural invention provided by the present invention. A socially coupled watershed hydrological drought assessment system is described below, including a natural one. A socially coupled watershed hydrological drought assessment system and a natural one described above Socially coupled watershed hydrological drought assessment methods can be used for mutual reference.

[0093] This invention provides a natural A socially coupled watershed hydrological drought assessment system is used to implement a natural [system / mechanism] provided in this invention. Socially coupled watershed hydrological drought assessment methods include: precipitation The runoff relationship variation diagnosis module is used to collect precipitation and runoff observation data of the target watershed, and to control the precipitation at the cross-sections of the collected precipitation and runoff observation data of the target watershed. Diagnosis of variations in runoff relationships; The drought index calculation module is used to calculate the drought index, which includes the meteorological drought index and the hydrological drought index that takes into account the impact of reservoir operation. Based on nature A socially coupled drought threshold calculation module is used to calculate drought thresholds based on natural factors. The drought threshold of social coupling, based on nature Socially coupled drought thresholds include standardized meteorological drought thresholds and standardized hydrological drought thresholds; The watershed drought risk assessment module is used to assess precipitation based on control sections. The results of runoff relationship variation diagnosis, combined with the drought index, are used to calculate the response time of hydrological drought to meteorological drought, based on natural... A socially coupled drought threshold is proposed, which estimates the conditional probability and propagation threshold of hydrological drought based on the Copula function, and assesses the risk of drought in the watershed based on the conditional probability and propagation threshold of hydrological drought.

[0094] Figure 8 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 8 As shown, the electronic device may include a processor 810, a communication interface 820, a memory 830, and a communication bus 840. The processor 810, communication interface 820, and memory 830 communicate with each other via the communication bus 840. The processor 810 can call logical instructions from the memory 830 to execute a natural... A socially coupled watershed hydrological drought assessment method, which includes: S1, collect precipitation and runoff observation data of the target watershed, and control the precipitation at the cross-sections based on the collected precipitation and runoff observation data of the target watershed. Diagnosis of variations in runoff relationships; S2, calculate the drought index, which includes the meteorological drought index and the hydrological drought index considering the impact of reservoir operation; S3, calculation based on nature The drought threshold of social coupling, based on nature Socially coupled drought thresholds include standardized meteorological drought thresholds and standardized hydrological drought thresholds; S4, based on control section precipitation The results of runoff relationship variation diagnosis, combined with the drought index, are used to calculate the response time of hydrological drought to meteorological drought, based on natural... A socially coupled drought threshold is proposed, which estimates the conditional probability and propagation threshold of hydrological drought based on the Copula function, and assesses the risk of drought in the watershed based on the conditional probability and propagation threshold of hydrological drought.

[0095] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: USB flash drive, portable hard drive, read-only memory (ROM), etc. Various media that can store program code, such as only memory, random access memory (RAM), magnetic disks or optical disks.

[0096] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to perform a natural function provided by the methods described above. A socially coupled watershed hydrological drought assessment method, which includes: S1, collect precipitation and runoff observation data of the target watershed, and control the precipitation at the cross-sections based on the collected precipitation and runoff observation data of the target watershed. Diagnosis of variations in runoff relationships; S2, calculate the drought index, which includes the meteorological drought index and the hydrological drought index considering the impact of reservoir operation; S3, calculation based on nature The drought threshold of social coupling, based on nature Socially coupled drought thresholds include standardized meteorological drought thresholds and standardized hydrological drought thresholds; S4, based on control section precipitation The results of runoff relationship variation diagnosis, combined with the drought index, are used to calculate the response time of hydrological drought to meteorological drought, based on natural... A socially coupled drought threshold is proposed, which estimates the conditional probability and propagation threshold of hydrological drought based on the Copula function, and assesses the risk of drought in the watershed based on the conditional probability and propagation threshold of hydrological drought.

[0097] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the natural methods described above. A socially coupled watershed hydrological drought assessment method, which includes: S1, collect precipitation and runoff observation data of the target watershed, and control the precipitation at the cross-sections based on the collected precipitation and runoff observation data of the target watershed. Diagnosis of variations in runoff relationships; S2, calculate the drought index, which includes the meteorological drought index and the hydrological drought index considering the impact of reservoir operation; S3, calculation based on nature The drought threshold of social coupling, based on nature Socially coupled drought thresholds include standardized meteorological drought thresholds and standardized hydrological drought thresholds; S4, based on control section precipitation The results of runoff relationship variation diagnosis, combined with the drought index, are used to calculate the response time of hydrological drought to meteorological drought, based on natural... A socially coupled drought threshold is proposed, which estimates the conditional probability and propagation threshold of hydrological drought based on the Copula function, and assesses the risk of drought in the watershed based on the conditional probability and propagation threshold of hydrological drought.

[0098] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0099] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A kind of nature A socially coupled watershed hydrological drought assessment method is characterized by, Includes the following steps: Collect precipitation and runoff observation data for the target watershed, and then use the collected precipitation and runoff observation data at control sections to analyze the precipitation data. Diagnosis of variations in runoff relationships; The drought index is calculated, which includes the meteorological drought index and the hydrological drought index that takes into account the impact of reservoir operation. Computation based on nature The drought threshold of social coupling, based on nature Socially coupled drought thresholds include standardized meteorological drought thresholds and standardized hydrological drought thresholds; Based on control section precipitation The results of runoff relationship variation diagnosis, combined with the drought index, are used to calculate the response time of hydrological drought to meteorological drought, based on natural... A socially coupled drought threshold is proposed, which estimates the conditional probability and propagation threshold of hydrological drought based on the Copula function, and assesses the risk of drought in the watershed based on the conditional probability and propagation threshold of hydrological drought.

2. The natural one according to claim 1 A socially coupled watershed hydrological drought assessment method is characterized by, The methods for collecting precipitation and runoff observation data of the target watershed include: Runoff observation data from hydrological stations in the target watershed were collected to obtain the monthly runoff sequence of the control section; precipitation observation data from meteorological stations in the target watershed were collected, and interpolation techniques were used to generate the corresponding monthly precipitation sequence for the runoff area of ​​the control section.

3. The natural one according to claim 2 A socially coupled watershed hydrological drought assessment method is characterized by, The control section precipitation data is obtained by collecting precipitation and runoff observation data of the target watershed. Methods for diagnosing runoff variation and identifying abrupt change points include: By selecting the sliding step size and sliding window length of the monthly precipitation and runoff series, the Copula function is used to calculate the precipitation for each sliding window sequentially. Runoff correlation coefficient; The optimal Copula function for each sliding window is determined sequentially, and the precipitation is obtained as the sliding step size increases. Runoff correlation coefficient sequence; The obtained precipitation Change point diagnosis was performed on the runoff correlation coefficient sequence to obtain abrupt change points.

4. The natural one according to claim 3 A socially coupled watershed hydrological drought assessment method is characterized by, The obtained precipitation Methods for identifying abrupt change points using runoff correlation coefficient sequences include: Construct a rank-based cumulative sum statistic to quantify precipitation. Two precipitation periods before and after any candidate point in the runoff correlation coefficient sequence The overall difference in runoff correlation coefficients yields the absolute values ​​of the statistics corresponding to all change points; Calculate the probability of the maximum absolute value of the change point statistic occurring, and compare the probability of the maximum absolute value of the change point statistic occurring with a preset significance level threshold: if the probability of the maximum absolute value of the change point statistic occurring is less than or equal to the preset significance level threshold, then the time corresponding to the maximum absolute value of the change point statistic is determined to be a change point.

5. The natural one according to claim 4 A socially coupled watershed hydrological drought assessment method is characterized by, The method for calculating the drought index includes: The cumulative probability distribution of monthly precipitation is obtained by fitting the monthly precipitation sequence; The standardized precipitation index (SPI), or meteorological drought index, is calculated by approximate algorithm after normalizing the monthly cumulative probability distribution. The reservoir index was used as a covariate, and the cumulative probability distribution of monthly runoff at the control section was calculated using the GAMLSS model. The cumulative probability distribution of monthly runoff was then normalized to obtain the non-uniform standardized runoff index NSSI, which is the hydrological drought index that takes into account the impact of reservoir scheduling.

6. The natural one according to claim 5 A socially coupled watershed hydrological drought assessment method is characterized by, The method for calculating the hydrological drought index, which takes into account the impact of reservoir operation, includes the following sub-steps: Determine the non-consistent alternative combinations of runoff probability distribution models and covariates. The probability distribution parameters include location parameters, scale parameters, and shape parameters. At least one of the location parameters and scale parameters contains a covariate, which includes the reservoir index. By inputting the runoff probability distribution model and the non-consistent alternative combinations of covariates, as well as the cumulative runoff sequences for each month, into the GAMLSS model, the optimal non-consistent combination of the runoff probability distribution model and covariates is determined, and the monthly cumulative runoff probability distribution at the control section is obtained. The non-uniform standardized runoff index (NSSI) is obtained by normalizing the cumulative probability distribution of monthly runoff at the control section, which is a hydrological drought index that takes into account the impact of reservoir scheduling.

7. The natural one according to claim 1 A socially coupled watershed hydrological drought assessment method is characterized by, The basis for precipitation control sections Methods for calculating the response time of hydrological drought to meteorological drought, based on runoff relationship variation diagnostic results and combined with the drought index, include: For undiagnosed precipitation For the control sections where runoff relationship changes abruptly, causal analysis is performed using the entire sequences of meteorological drought index and hydrological drought index at the control sections to determine the response time of hydrological drought to meteorological drought. For diagnosing precipitation The control section of the runoff relationship abrupt change point is used as the boundary to divide the meteorological drought index and hydrological drought index into two time periods before and after the change point for causal relationship analysis, and to determine the response time of hydrological drought to meteorological drought before and after the change point.

8. The natural one according to claim 7 A socially coupled watershed hydrological drought assessment method is characterized by, The methods for conducting causal relationship analysis include: The correlation coefficient between the meteorological drought index and the hydrological drought index was calculated using a convergent cross-mapping algorithm with time lag, and plotted on the time lag. In the correlation coefficient plane coordinate system, the x-coordinate corresponding to the highest point in the obtained scatter plot is taken as the response time of hydrological drought to meteorological drought.

9. The natural one according to claim 8 A socially coupled watershed hydrological drought assessment method is characterized by, The basis is based on nature Socially coupled drought thresholds and methods for estimating the conditional probability of hydrological drought occurrence based on Copula functions include: Based on nature The socially coupled drought threshold was used as a cutoff threshold to extract meteorological drought events and hydrological drought events from the drought index sequence, respectively. Considering the response time of hydrological drought to meteorological drought, the extracted meteorological drought events and hydrological drought events are matched to obtain meteorological drought data. For hydrological drought event pairs, extract the corresponding drought characteristic variables for each drought event pair; The drought characteristic variables obtained by matching are regarded as random variables. The marginal probability distribution parameters are fitted, and the most suitable marginal probability distribution is obtained according to the runoff probability distribution model and the optimal combination of covariates with non-consistent covariates. The Vine Copula function is used to process the multivariate joint distribution of drought characteristic variables. By decomposing them into paired Copula structures, the drought characteristic variables include meteorological drought duration, meteorological drought intensity, meteorological drought peak, and hydrological drought intensity. Based on the cumulative probability of the matched meteorological drought intensity, hydrological drought events are divided into four levels: mild, moderate, severe, and extreme. When the meteorological drought characteristic variables exceed the preset threshold, the conditional probability of triggering severe and extreme hydrological drought events is estimated in each sub-interval within the specified threshold range.

10. A kind of nature A socially coupled watershed hydrological drought assessment system is characterized by, For implementing claim 1 Any one of the following 9 natural Socially coupled watershed hydrological drought assessment methods include: precipitation The runoff relationship variation diagnosis module is used to collect precipitation and runoff observation data of the target watershed, and to control the precipitation at the cross-sections of the collected precipitation and runoff observation data of the target watershed. Diagnosis of variations in runoff relationships; The drought index calculation module is used to calculate the drought index, which includes the meteorological drought index and the hydrological drought index that takes into account the impact of reservoir operation. Based on nature A socially coupled drought threshold calculation module is used to calculate drought thresholds based on natural factors. The drought threshold of social coupling, based on nature Socially coupled drought thresholds include standardized meteorological drought thresholds and standardized hydrological drought thresholds; The watershed drought risk assessment module is used to control cross-sectional precipitation. The results of runoff relationship variation diagnosis, combined with the drought index, are used to calculate the response time of hydrological drought to meteorological drought, based on natural... A socially coupled drought threshold is proposed, which estimates the conditional probability and propagation threshold of hydrological drought based on the Copula function, and assesses the risk of drought in the watershed based on the conditional probability and propagation threshold of hydrological drought.

Images