A hydrological drought classification early warning method based on water shortage threshold

By calculating the spread time and difference in precipitation and evaporation between meteorological drought to hydrological drought, and combining the Copula function to construct a joint probability distribution, the problems of inconvenient monitoring and low prediction accuracy of hydrological drought warning in the existing technology are solved, and more accurate and practical hydrological drought warning.

CN116611542BActive Publication Date: 2024-12-03JIANGXI ACAD OF WATER SCI & ENG
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310311461.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-03-28
Publication Date
2024-12-03
Estimated Expiration
2043-03-28

AI Technical Summary

Technical Problem

The existing hydrological drought early warning methods have problems such as inconvenient monitoring and low prediction accuracy. Especially because the duration and intensity of meteorological drought are difficult to monitor, the threshold indicators of hydrological drought events are inaccurate, and the impact of human activities and climate factors is complex, resulting in excessive matching. and frequent warnings.

Method used

By calculating the propagation time of meteorological drought to hydrological drought, identifying and matching meteorological drought and hydrological drought events, and computing precipitation shortage sequences based on the difference between precipitation and evaporation, a two-dimensional joint probability distribution was constructed using the Copula function to estimate the duration and intensity of hydrological droughts. The water shortage threshold eliminates the impact of human activities and improves the accuracy of the threshold.

Benefits of technology

It improves the accuracy and practicality of hydrological drought warnings, the rationality of precipitation shortage sequences, enhances the convenience of monitoring and early warnings, reduces the phenomenon of over-matching and frequent early warnings, and enhances the ability to prevent droughts.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116611542B_ABST
    Figure CN116611542B_ABST
Patent Text Reader

Abstract

The present invention discloses a hydrological drought graded early warning method based on water shortage threshold, comprising the following steps: calculating the standardized precipitation evapotranspiration index (SPEI) and the standardized runoff index (SRI), determining the propagation time of meteorological drought to hydrological drought; identifying meteorological drought and hydrological drought events, and matching hydrological and meteorological drought events; calculating the precipitation shortage sequence, extracting the duration and intensity of hydrological drought; selecting appropriate marginal distribution functions to fit the precipitation shortage sequence and the duration and intensity of hydrological drought respectively, and estimating their parameters; using the Copula function to construct the two-dimensional joint probability distribution function of the precipitation shortage sequence and the duration and intensity of hydrological drought respectively, and estimating the function parameters; using the conditional probability distribution expression to calculate the water shortage threshold of the duration and intensity of hydrological drought. The present invention proposes a hydrological drought early warning method to achieve more accurate and easy-to-monitor hydrological drought early warning, which can provide an important scientific basis for active drought prevention and early warning.
Need to check novelty before this filing date? Find Prior Art

Description

Technical field:

[0001] The present invention relates to the technical field of drought management and response, and belongs to the technical field of hydrological drought research, and particularly to a hydrological drought graded early warning method based on a water shortage threshold. Background technology:

[0002] Meteorological drought caused by long-term precipitation deficit may further lead to insufficient runoff, lakes and reservoirs through the water cycle process, which may induce hydrological drought. Once a hydrological drought occurs, it will have a direct impact on residents' domestic water use, agricultural irrigation water use and ecological environment water use. Severe hydrological drought will cause huge economic losses. Therefore, conducting hydrological drought early warning research can effectively improve drought prevention capabilities and reduce drought losses.

[0003] Hydrological drought is usually caused by meteorological drought, and the two are closely related, which provides the possibility of warning hydrological drought through meteorological drought. At present, the commonly used hydrological drought warning method is to establish a response model between the characteristic values ​​of meteorological drought and hydrological drought. However, related research has certain limitations in hydrological drought warning. On the one hand, there is a lack of exploration of the attribution of hydrological drought events, and the hydrological drought events dominated by human disturbances are included in the analysis, resulting in over-matching of drought event samples, affecting the accuracy of the threshold index; on the other hand, the duration and intensity of meteorological drought are not easy to monitor as the threshold of the warning index of hydrological drought, and it is inconvenient to apply it in practice. Through the above analysis, the present invention proposes to refine the propagation process of meteorological drought to hydrological drought, and to calculate the threshold value of hydrological drought with actual precipitation shortage as an indicator. The present invention enriches and develops the existing drought warning methods, which has important practical significance for the relevant departments to deploy in advance, actively fight drought, and reduce the social and economic losses caused by drought.

[0004] 1. Technical solution of the prior art 1: Taking the humid version of the humid area as an example, the prediction and early warning method of the transmission of meteorological drought to hydrological drought is discussed. This method uses SPI and the standardized runoff index (SSI) to obtain the characteristics of meteorological drought and hydrological drought, and uses linear or nonlinear fitting to construct a response model of hydrological drought to meteorological drought. The fitting model is used to obtain the duration and intensity characteristics of meteorological drought that triggers hydrological drought.

[0005] Disadvantages of the first prior art: The method proposed in this scheme uses the duration and intensity characteristics of meteorological drought as the threshold indicators of hydrological drought, but the duration and intensity of drought are not observed values. When used as thresholds, they need to be calculated multiple times, which is difficult to obtain in actual monitoring, resulting in inconvenience in monitoring.

[0006] 2. The technical solution of the second prior art explores the use of the difference between the precipitation during the drought period and the multi-year average precipitation to define the precipitation deficit value. Using the drought propagation time t, the precipitation deficit value is obtained by moving forward t months from the start month of the hydrological drought, and then the precipitation deficit sequence matching the number of hydrological drought events is obtained. Then, the copula function and conditional probability are used to construct a joint distribution to obtain the water deficit threshold of the hydrological drought.

[0007] Disadvantages of the second existing technology: This method uses the precipitation shortage value as the threshold indicator, although it is more conducive to monitoring and early warning than the method one that uses drought characteristics as the threshold. However, this method determines the precipitation shortage value by moving the starting month of the hydrological drought forward t months. In fact, the propagation of drought is affected by many factors such as watershed regulation, human activities, and climate variables, which makes the way meteorological drought propagates to hydrological drought complex and not one-to-one. This method will cause over-matching of hydrological drought events; at the same time, this method describes the precipitation shortage as the difference between the precipitation during the drought period and the multi-year average precipitation, while ignoring the impact of evapotranspiration during the drought period, which will cause the precipitation shortage threshold to be seriously underestimated, and frequent warnings may occur. Therefore, this scheme has certain limitations in the actual application of hydrological drought graded early warning. Summary of the invention:

[0008] The purpose of the present invention is to overcome the deficiencies of the prior art and provide a hydrological drought classification early warning method based on water shortage thresholds, which can effectively make up for the problems of inconvenience in using existing technical indicators for monitoring and early warning, and low prediction accuracy.

[0009] In order to solve the above technical problems, the technical solution adopted by the present invention is as follows: the hydrological drought graded early warning method based on water shortage threshold comprises the following steps: S1: calculating SPEI and SRI indexes to determine the propagation time of meteorological drought to hydrological drought; S2: identifying meteorological drought and hydrological drought events, and matching hydrological and meteorological drought events; S3: calculating precipitation shortage sequence, extracting hydrological drought duration and intensity; S4: selecting appropriate marginal distribution function according to hydrological drought duration, intensity and precipitation shortage sequence data in S3, and estimating its parameters; S5: using Copula function to construct two-dimensional joint probability distribution functions of precipitation shortage sequence and hydrological drought duration and intensity respectively, and estimating function parameters; S6: using conditional probability distribution expression to calculate the water shortage threshold of hydrological drought duration and intensity.

[0010] As a preferred method, S1 is as follows: the potential evapotranspiration is calculated by using the Penman-Monteith formula considering the physical process through the obtained monthly meteorological data. The difference between precipitation and evapotranspiration is calculated to obtain the water surplus and shortage sequence, and the sequence is normalized and processed to obtain the SPEI value; the SRI index is calculated by the obtained monthly runoff data. Similarly, the runoff data is fitted using the Gamma distribution, and then the probability sequence is transformed into the standard normalization consistent with the SPEI to obtain the SRI sequence.

[0011] Preferably, S1 is specifically as follows: SRI (SRI-1) at a one-month time scale and SPEI (SPEI-i) at multiple time scales are calculated to determine the propagation time of meteorological drought to hydrological drought.

[0012] t=τ(max(P1,P2,……,Pi))

[0013] Where t is the time for meteorological drought to propagate into hydrological drought; Pi is the Pearson correlation coefficient between SPEI and SRI-1 at the time scale of i months; τ is the calculation symbol, i.e., the time scale i corresponding to Pi is obtained.

[0014] As the preferred S2, the specific method is as follows: using the run theory to identify meteorological drought and hydrological drought events, using the drought propagation time t to match meteorological drought and hydrological drought events, and determining the water shortage response duration of the hydrological drought.

[0015] When meteorological drought matches hydrological drought one-to-one:

[0016] T=Me-Ms (1)

[0017] Where Me is the month when the meteorological drought ends, and Ms is the month when the meteorological drought begins.

[0018] When a meteorological drought triggers two adjacent hydrological droughts:

[0019] T1=He1-Ms (2)

[0020] T2=Hs2-Ms (3)

[0021] Where T1 is the duration of water shortage response corresponding to the first hydrological drought, T2 is the duration of water shortage response corresponding to the second hydrological drought, He1 is the month when the first hydrological drought ends, and Hs2 is the month when the second hydrological drought begins.

[0022] When multiple meteorological droughts trigger hydrological droughts:

[0023] T=Men-Ms1 (4)

[0024] Where Men is the month when the last meteorological drought event ended, and Ms1 is the month when the first meteorological drought began.

[0025] The preferred S3 is as follows: the duration and intensity characteristics of hydrological drought are extracted using the run theory, the water shortage response duration of each hydrological drought is determined, and then the precipitation shortage within the water shortage response duration corresponding to each hydrological drought event is calculated to obtain the precipitation shortage sequence.

[0026]

[0027] Where CPD is the precipitation deficit, T is the water deficit response duration, P is the precipitation, and PET is the evapotranspiration.

[0028] The preferred S4 is as follows: Gamma (Gam) distribution, Weibull (Wbl) distribution, lognormal (Logn) distribution, and Generalized extreme value (Gev) distribution are used to fit the hydrological drought duration, intensity, and precipitation shortage series, respectively. The Kolmogorov-Smirnov (KS) test method is used to evaluate the optimal marginal distribution function, and the maximum likelihood method (MLE) is used to evaluate the parameters in the distribution function.

[0029] As the preferred S5, the specific details are as follows: using copula function to construct the two-dimensional joint distribution function of the hydrological drought duration and precipitation shortage series, the hydrological drought severity and precipitation shortage series, the copula function type selects Frank copula, Gumbel copula, Clayton copula, and the Akashi information criterion (AIC) and the squared Euclidean distance (d 2 ) to select the best copula function type and use the MLE method to evaluate the parameters in the copula function.

[0030] The preferred S6 is as follows: based on the joint distribution and conditional distribution formula of Copula, the precipitation shortage condition thresholds corresponding to hydrological droughts of different drought levels are calculated; with precipitation shortage as the condition and the duration and intensity of hydrological droughts of different levels as the target, the water shortage thresholds of hydrological drought events of different levels are determined; the confidence level of the meteorological condition threshold triggered by the hydrological drought is determined to be 95%, that is, when the conditional probability that the meteorological condition triggers the hydrological event is greater than or equal to 0.95, the corresponding cumulative precipitation shortage is regarded as the meteorological condition threshold that triggers the hydrological drought event of this level.

[0031]

[0032] where x and y are the marginal cumulative distributions of two random variables X and Y; x(u) and y(v) are the cumulative probabilities of X≤u and Y≤v respectively; and C(x(u), y(v)) is the joint cumulative probability of X≤u and Y≤v.

[0033] After adopting the above technical solution, the present invention has the following beneficial effects compared with the prior art:

[0034] (1) The hydrological drought graded warning calculation method based on water shortage thresholds calculates the propagation time from meteorological drought to hydrological drought, and uses the drought propagation time to identify and match meteorological drought and hydrological drought one by one. The meteorological and hydrological drought sequence determined by the one-to-one matching method eliminates hydrological drought events caused by human activities, and the accuracy of the precipitation shortage sequence obtained is better than the precipitation shortage sequence directly determined by the drought propagation time.

[0035] (2) The meteorological drought events that trigger hydrological drought are determined by drought event matching, and the meteorological drought precipitation deficit value is calculated based on the difference between precipitation and potential evapotranspiration to obtain the precipitation deficit sequence. The obtained precipitation deficit sequence takes into account the relationship between precipitation and evaporation, improves the rationality of the precipitation deficit sequence calculation, and thus improves the accuracy of the hydrological drought water deficit threshold.

[0036] (3) Compared with warning based on the duration and intensity of meteorological drought, using the precipitation deficit of hydrological drought as the threshold indicator is more practical and convenient for monitoring and early warning. Description of the drawings:

[0037] Figure 1 A flow chart of a hydrological drought classification early warning method based on water shortage thresholds provided in an embodiment of the present invention.

[0038] Figure 2 A correlation diagram between SRI-1 and SPEI on a time scale of 1-12 months provided in an embodiment of the present invention.

[0039] Figure 3 A schematic diagram of the run-length theory provided for an embodiment of the present invention.

[0040] Figure 4 A diagram of hydrological drought and water shortage thresholds of different degrees provided by an embodiment of the present invention. Specific implementation method:

[0041] The present invention is described in detail below in conjunction with embodiments, but the present invention is not limited to these embodiments.

[0042] like Figure 1 As shown, the present invention is a hydrological drought classification early warning method based on water shortage threshold, which includes the following steps:

[0043] S1. Collect meteorological and hydrological data, and calculate the standardized precipitation evapotranspiration index (SPEI) and standardized runoff index (SRI). In this embodiment, the monthly meteorological and runoff data from 1960 to 2018 at Julongtan Station, Xiashan Station and Hanlinqiao Station at the source of Ganjiang River were selected for research. The SRI on a 1-month time scale and the SPEI index series on a 1-12-month time scale were calculated. The propagation time of meteorological drought to hydrological drought was determined based on the time lag effect; combined with Figure 2 As shown, correlation coefficients were calculated between SRI-1 and 1-12 month scale SPEI.

[0044] The details are as follows: The potential evapotranspiration is calculated using the Penman-Monteith formula that takes into account the physical process through the monthly meteorological data obtained. The difference between precipitation and evapotranspiration is calculated to obtain the water surplus and shortage sequence, and the sequence is normalized and finally the SPEI value is obtained. The SRI index is calculated using the monthly runoff data obtained. Similarly, the runoff data is fitted using the Gamma distribution, and then the probability sequence is transformed into the standard normalization consistent with the SPEI to obtain the SRI sequence.

[0045] The details are as follows: The SRI on a one-month time scale (SRI-1) and the SPEI on multiple time scales (SPEI-i) are calculated to determine the propagation time of meteorological drought to hydrological drought.

[0046] t=τ(max(P1,P2,……,Pi))

[0047] Where t is the time for meteorological drought to propagate into hydrological drought; Pi is the Pearson correlation coefficient between SPEI and SRI-1 at the time scale of i months; τ is the calculation symbol, i.e., the time scale i corresponding to Pi is obtained.

[0048] S2. Use the travel theory to identify meteorological drought and hydrological drought events, and use the drought propagation time to match the meteorological drought that causes the hydrological drought event. The details are as follows:

[0049] The travel theory is used to identify meteorological drought and hydrological drought events, and the drought propagation time t is used to match meteorological drought and hydrological drought events to determine the water shortage response duration of hydrological drought.

[0050] When meteorological drought matches hydrological drought one-to-one:

[0051] T=Me-Ms (1)

[0052] Where Me is the month when the meteorological drought ends, and Ms is the month when the meteorological drought begins.

[0053] When a meteorological drought triggers two adjacent hydrological droughts:

[0054] T1=He1-Ms (2)

[0055] T2=Hs2-Ms (3)

[0056] Where T1 is the duration of water shortage response corresponding to the first hydrological drought, T2 is the duration of water shortage response corresponding to the second hydrological drought, He1 is the month when the first hydrological drought ends, and Hs2 is the month when the second hydrological drought begins.

[0057] When multiple meteorological droughts trigger hydrological droughts:

[0058] T=Men-Ms1 (4)

[0059] Where Men is the month when the last meteorological drought event ended, and Ms1 is the month when the first meteorological drought began.

[0060] like Figure 3 As shown, it is a three-parameter run theory method. The thresholds used in this example are R0=0, R1=-0.5, and R2=-1. The duration of hydrological drought is the number of months the drought process lasts, and the drought intensity refers to the cumulative value of SPEI of the drought process.

[0061] S3. Use the run theory to extract the duration and intensity of hydrological droughts, and calculate the precipitation shortage data sequence of the matched meteorological drought events. Specifically, use the run theory to extract the duration and intensity characteristics of hydrological droughts, determine the water shortage response duration of each hydrological drought, and then calculate the precipitation shortage within the water shortage response duration corresponding to each hydrological drought event to obtain the precipitation shortage sequence.

[0062]

[0063] Where CPD is the precipitation deficit, T is the water deficit response duration, P is the precipitation, and PET is the evapotranspiration.

[0064] S4. Based on the hydrological drought and precipitation shortage data series, specifically, the Gam distribution, Wbl distribution, Logn distribution, and Gev distribution were used to fit the hydrological drought duration, intensity, and precipitation shortage series, respectively. The Kolmogorov-Smirnov (KS) test method was used to evaluate the optimal marginal distribution function, and the maximum likelihood method (MLE) was used to evaluate the parameters in the distribution function.

[0065] Table 1 Marginal distribution function fitting

[0066]

[0067] The bracketed values ​​in Table 1 indicate the optimal marginal distribution.

[0068] S5. Use copula function to construct the two-dimensional joint distribution function of hydrological drought duration and precipitation shortage series, hydrological drought intensity and precipitation shortage series. The copula function type selects Frank copula, Gumbel copula, and Clayton copula. The Akagi Information Criterion (AIC) and the squared Euclidean distance (d 2 ) to select the best copula function type and use the MLE method to evaluate the parameters in the copula function.

[0069] Table 2 Joint distribution function

[0070]

[0071] The values ​​in brackets in Table 2 are the optimal copula function types.

[0072] It can be seen from Table 2 that the optimal copula function type for water shortage and hydrological drought duration and intensity at Xiashan station is Frank copula function, while Julongtan and Hanlinqiao stations are Gumbel copula function type.

[0073] S6. The hydrological drought water shortage threshold is calculated using the conditional probability distribution. The meteorological triggering thresholds of hydrological drought events under the conditions of moderate drought, severe drought and extreme drought are calculated, and their cumulative probability critical values ​​are 0.5, 0.75 and 0.9 respectively. In this example, the confidence level of the meteorological condition threshold triggered by hydrological drought is 95%, that is, when the conditional probability of meteorological conditions triggering the hydrological event is greater than or equal to 0.95, the corresponding cumulative precipitation shortage is regarded as the meteorological condition threshold that triggers the hydrological drought event of this level.

[0074]

[0075] where x and y are the marginal cumulative distributions of two random variables X and Y; x(u) and y(v) are the cumulative probabilities of X≤u and Y≤v respectively; and C(x(u), y(v)) is the joint cumulative probability of X≤u and Y≤v.

[0076] like Figure 3 When the cumulative probability of hydrological drought duration and intensity reaches 0.95, the corresponding water shortage is the water shortage threshold under the corresponding drought degree.

[0077] The hydrological drought water shortage threshold calculation method of the present invention improves the accuracy of hydrological drought early warning, and taking water shortage as an early warning indicator has stronger practical significance, has relatively large value for drought early warning, and has good application effect.

[0078] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods or computer program products. The present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media including computer-usable program code.

[0079] The present application is described with reference to the flowchart or block diagram of the method, device (system), and computer program product according to the embodiment of the present application. It should be understood that each process in the flowchart or block diagram, and the combination of the processes in the flowchart or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine.

[0080] The above are only a few embodiments of the present application, and are not intended to limit the present application in any form. However, the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned embodiments are only illustrative, not restrictive. Professionals and technicians familiar with the field, without departing from the scope of protection of the claims of the present invention, any changes or modifications made are equivalent to equivalent implementation cases and fall within the scope of the technical solution.

Claims

1. A hydrological drought classification early warning method based on water shortage threshold, characterized in that: The following steps are involved: S1: Calculate the Standardized Precipitation Evapotranspiration Index (SPEI) and Standardized Runoff Index (SRI) to determine the propagation time of meteorological drought to hydrological drought; S2: Identify meteorological drought and hydrological drought events, and match hydrological and meteorological drought events; S3: Calculate precipitation shortage sequence, extract hydrological drought duration and intensity; S4: According to the hydrological drought duration, intensity and precipitation shortage sequence data in S3, select appropriate marginal distribution function and estimate its parameters; S5: Use Copula function to construct two-dimensional joint probability distribution function of precipitation shortage sequence and hydrological drought duration and intensity respectively, and estimate function parameters; S6: Use conditional probability distribution expression to calculate the water shortage threshold of hydrological drought duration and intensity; Among them, S1 calculates the SPEI and SRI indexes through the monthly meteorological and runoff data, calculates the SRI-1 on a one-month time scale and the SPEI-i on multiple time scales, and determines the propagation time of meteorological drought to hydrological drought: t=τ(max (P1,P2,…,Pi)) (1) Where t is the time for meteorological drought to propagate to hydrological drought; Pi is the Pearson correlation coefficient between SPEI and SRI-1 at the time scale of i months; τ is the calculation symbol, that is, the time scale i corresponding to Pi is obtained; S2 uses the travel theory to identify meteorological and hydrological drought events, uses the drought propagation time t to match meteorological drought and hydrological drought events, and determines the duration of hydrological drought water shortage response; the specific steps are as follows: When meteorological drought matches hydrological drought one-to-one: T=Me-Ms (1) In the formula, Me is the month when the meteorological drought ends, and Ms is the month when the meteorological drought begins; When a meteorological drought triggers two adjacent hydrological droughts: T1=He1-Ms (2) T2=Hs2-Ms (3) Where T1 is the duration of water shortage response corresponding to the first hydrological drought, T2 is the duration of water shortage response corresponding to the second hydrological drought, He1 is the month when the first hydrological drought ends, and Hs2 is the month when the second hydrological drought begins; When multiple meteorological droughts trigger hydrological droughts: T=Me n -Ms1 (4) In the formula, Me n Ms is the month when the last meteorological drought event ended, and Ms1 is the month when the first meteorological drought began; S3 uses the run theory to extract the duration and intensity characteristics of hydrological drought, determines the water shortage response duration of each hydrological drought event after matching, and then calculates the precipitation shortage within the water shortage response duration corresponding to each hydrological drought event to obtain the precipitation shortage sequence; Where CPD is the precipitation deficit, T is the water deficit response duration, P is the precipitation, and PET is the evapotranspiration.

2. The hydrological drought classification early warning calculation method based on water shortage threshold according to claim 1 is characterized by: S4 used Gamma distribution, Weibull distribution, lognormal distribution, and Generalized extreme value distribution to fit the hydrological drought duration, intensity, and precipitation shortage series, respectively. The Kolmogorov-Smirnov (KS) test method was used to evaluate the best marginal distribution function, and the maximum likelihood method (MLE) was used to evaluate the parameters in the distribution function.

3. The hydrological drought classification early warning calculation method based on water shortage threshold according to claim 1 is characterized by: S5 uses copula function to construct the two-dimensional joint distribution function of precipitation shortage series and hydrological drought duration and intensity respectively. The copula function type selects Frank copula, Gumbel copula and Clayton copula. The Akagi Information Criterion (AIC) and the squared Euclidean distance (d 2 ) to select the best copula function type and use the MLE method to evaluate the parameters in the copula function.

4. The hydrological drought classification early warning calculation method based on water shortage threshold according to claim 1 is characterized by: S6 calculates the water shortage thresholds of different levels of hydrological drought duration and intensity based on the Copula function and conditional probability distribution formula; Taking precipitation shortage as the condition and the duration and intensity of hydrological drought of different levels as the target, the water shortage thresholds of hydrological drought events of different levels are determined; The confidence level of the meteorological condition threshold triggering hydrological drought is determined to be 95%, that is, when the conditional probability of meteorological conditions triggering this hydrological drought event is greater than or equal to 0.95, the corresponding cumulative precipitation deficit is regarded as the meteorological condition threshold triggering this level of hydrological drought event: where x and y are the marginal cumulative distributions of two random variables X and Y; x(u) and y(v) are the cumulative probabilities of X≤u and Y≤v respectively; and C(x(u), y(v)) is the joint cumulative probability of X≤u and Y≤v.