Multi-type drought propagation process analysis method, device, medium and equipment

Abstract

The invention discloses a multi-type drought propagation process analysis method and device, a medium and equipment, and relates to the field of drought propagation data processing, and the method comprises the steps: obtaining meteorological, hydrological, agricultural and underground water drought data of a to-be-detected region under different time scales, obtaining meteorological-hydrological, meteorological-agricultural and meteorological-underground water drought propagation time, and calculating the drought propagation time of the to-be-detected region according to the meteorological-hydrological, meteorological-agricultural and meteorological-underground water drought propagation time; determining a drought propagation path of meteorology-hydrology-agriculture-groundwater drought; the method comprises the following steps: constructing a bivariate joint distribution function to obtain the probability of causing drought risks under different meteorological drought levels, carrying out iteration on a meteorological drought index within a certain range with a fixed sliding step length, determining a drought triggering threshold, and obtaining the drought resistance of a to-be-detected region according to the change of the triggering threshold. According to the method, the drought propagation path is determined, and the drought trigger threshold value is combined, so that the drought resistance of a certain region can be accurately predicted.

Description

Technical Field

[0001] The present invention relates to the field of drought propagation data processing, and in particular to a multi-type drought propagation process analysis method, device, medium and equipment. Background Art

[0002] Among many extreme climate events, drought, as the most common natural disaster in the world, has the characteristics of high frequency, long duration, wide range and large losses. A report released by the World Meteorological Organization (WMO) in 2021 showed that in the past half century, an average of one climate, weather or water-related disaster occurred every day, killing an average of 115 people every day, and drought was the disaster with the most casualties. However, as a sensitive area to climate change, my country is extremely vulnerable to the impact of climate change. In a changing environment, my country's drought will face more severe challenges.

[0003] Drought is a complex problem involving a wide range of areas. Different disciplines have different understandings and definitions of drought. Traditionally, drought types can be divided into four categories: meteorological, hydrological, agricultural, and socioeconomic droughts (Organization, 2006). Drought propagation is a complex process involving various components of the water cycle. However, current research on drought propagation focuses on two types of droughts, lacking systematic research that considers multiple types of droughts, making it difficult to accurately predict a region's resistance to drought. Summary of the invention

[0004] The present invention provides a multi-type drought propagation process analysis method to solve the above-mentioned problem existing in the prior art, that is, the problem of accurately predicting the resistance of a certain area to drought in the prior art. The present invention provides a multi-type drought propagation process analysis method, which comprises: Obtain meteorological, hydrological, agricultural and groundwater drought data at different time scales in the test area; Based on the meteorological, hydrological, agricultural and groundwater drought data at different time scales, the correlation between the meteorological drought index at different time scales and the hydrological drought index, agricultural drought index and groundwater drought index was determined, the propagation time of meteorological-hydrological, meteorological-agricultural and meteorological-groundwater droughts was obtained, and the drought propagation path of meteorological-hydrological-agricultural-groundwater drought was determined; Based on the drought propagation path of meteorological-hydrological-agricultural-groundwater drought, the non-parametric kernel density estimation method is used in combination with the connection function copula to fit the marginal distribution function corresponding to each drought index. By integrating the marginal distribution function, the joint distribution functions corresponding to meteorological-hydrological, hydrological-agricultural and agricultural-groundwater droughts are constructed respectively to determine the risk probability of causing the second-level drought in the tested area under different meteorological drought levels. According to the risk probability of drought, the meteorological drought index is iterated within a certain interval with a fixed sliding step size to obtain the drought risk probability result corresponding to each iteration. By presetting a drought risk probability value for hydrological drought, when the drought risk probability corresponding to each iteration is greater than or equal to the pre-set drought risk probability value, the interval of the meteorological drought index corresponding to this time is determined as the trigger threshold of drought in the test area. According to the change of the trigger threshold during the propagation process, the resistance of the test area to drought is obtained.

[0005] Optionally, the determination of the correlations between the meteorological drought index at different time scales and the hydrological drought index, the agricultural drought index and the groundwater drought index specifically includes: The Spearman correlation coefficient method was used to determine the maximum positive correlation between the meteorological drought index at different scales and the hydrological drought index, agricultural drought index and groundwater drought index.

[0006] Optionally, the non-parametric kernel density estimation method is combined with the connection function copula to fit the marginal distribution function corresponding to each drought index, specifically including: The marginal distribution function is obtained using the following formula: , in, is a random sample from an unknown distribution, is the window width, n is the number of samples, x is the data point that needs to be estimated, i is the random sample sequence number.

[0007] Optionally, the bivariate joint distribution function is constructed to determine the risk probability of causing a secondary drought in the tested area under different meteorological drought levels, specifically including: According to the bivariate joint distribution function and the Bayesian probability model, the conditional probability of hydrological drought relative to different levels of meteorological drought is analyzed to determine the risk probability of the next level of drought in the tested area under different levels of meteorological drought.

[0008] Optionally, the meteorological drought level specifically includes: Mild drought level, moderate drought level, severe drought level and extreme drought level.

[0009] The present invention provides a multi-type drought propagation process analysis device, comprising: The acquisition module is used to obtain meteorological, hydrological, agricultural and groundwater drought data at different time scales in the test area; The correlation analysis module is used to determine the correlation between the meteorological drought index at different time scales and the hydrological drought index, agricultural drought index and groundwater drought index respectively according to the meteorological, hydrological, agricultural and groundwater drought data at different time scales, obtain the propagation time of meteorological-hydrological, meteorological-agricultural and meteorological-groundwater droughts, and determine the drought propagation path of meteorological-hydrological-agricultural-groundwater droughts; A construction module is used to construct drought propagation paths based on meteorological-hydrological-agricultural-groundwater drought. The nonparametric kernel density estimation method is combined with the connection function copula to fit the marginal distribution functions corresponding to each drought index. By integrating the marginal distribution functions, the joint distribution functions corresponding to meteorological-hydrological, hydrological-agricultural and agricultural-groundwater droughts are constructed respectively to determine the risk probability of causing the next level of drought in the tested area under different meteorological drought levels. The determination module is used to iterate the meteorological drought index within a certain interval with a fixed sliding step size according to the risk probability of drought, and obtain the drought risk probability result corresponding to each iteration. By presetting a drought risk probability value of hydrological drought, when the drought risk probability corresponding to each iteration is greater than or equal to the pre-set drought risk probability value, the interval of the meteorological drought index corresponding to this time is determined as the trigger threshold of drought in the test area, and the resistance of the test area to drought is obtained according to the change of the trigger threshold during the propagation process.

[0010] The present invention provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned multi-type drought propagation process analysis method is implemented.

[0011] The present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned multi-type drought propagation process analysis method when executing the program.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention provides a method for analyzing the propagation process of multiple types of droughts. The method obtains meteorological, hydrological, agricultural and groundwater drought data at different time scales in the test area, determines the correlation between the meteorological drought index at different time scales and the hydrological drought index, the agricultural drought index and the groundwater drought index, and obtains the drought propagation time, thereby determining the drought propagation path of meteorological-hydrological-agricultural-groundwater drought; according to the drought propagation path, the joint distribution function corresponding to meteorological-hydrological, hydrological-agricultural and agricultural-groundwater droughts is constructed in combination with the connection function copula, and the risk probability of inducing a secondary drought is determined under different meteorological drought levels; finally, the drought risk probability is compared with a pre-set drought risk probability value to obtain a trigger threshold for inducing a secondary type of drought, and according to the change of the trigger threshold, the resistance of a certain area to drought is accurately predicted. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0014] Figure 1 A flowchart of a method for analyzing a multi-type drought propagation process provided by an embodiment of the present invention; Figure 2 A flow chart for calculating the propagation time of meteorological to hydrological drought provided by an embodiment of the present invention; Figure 3 A trigger threshold model framework diagram of a multi-type drought propagation process analysis method provided by an embodiment of the present invention; Figure 4 A schematic diagram of a computer device for a method for analyzing a multi-type drought propagation process provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] The technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0017] Figure 1 is a flow chart of a method for analyzing a multi-type drought propagation process provided by an embodiment of the present invention. Figure 1 As shown, this embodiment shows a method for analyzing multi-type drought propagation processes, including: S101: Obtain meteorological, hydrological, agricultural and groundwater drought data at different time scales in the test area.

[0018] S102: Based on the meteorological, hydrological, agricultural and groundwater drought data at different time scales, determine the correlation between the meteorological drought index at different time scales and the hydrological drought index, agricultural drought index and groundwater drought index, obtain the propagation time of meteorological-hydrological, meteorological-agricultural and meteorological-groundwater droughts, and determine the drought propagation path of meteorological-hydrological-agricultural-groundwater drought.

[0019] For example, SPEI, SRI, SSI, and GWSA-DSI may be used to describe the status of meteorological, hydrological, agricultural, and groundwater droughts, respectively.

[0020] Optionally, the Spearman correlation coefficient method is used to determine the maximum positive correlation between the meteorological drought index at different scales and the hydrological drought index, agricultural drought index and groundwater drought index, respectively.

[0021] like Figure 2 As shown in the figure, taking meteorological-hydrological drought as an example, the hydrological drought index SRI at a one-month scale and the SPEI at a 24-month scale are obtained, the Spearman correlation coefficient of the SRI at a one-month scale and the SPEI at 24 scales is calculated, and the maximum correlation coefficient and the corresponding SPEI time scale at this time are extracted. This time scale is the propagation time of meteorological to hydrological drought.

[0022] Similarly, the propagation times of meteorological-agricultural and meteorological-groundwater droughts can be obtained, and then the above three propagation times can be sorted to obtain different types of drought propagation paths.

[0023] S103: Based on the drought propagation path of meteorological-hydrological-agricultural-groundwater drought, the non-parametric kernel density estimation method is used in combination with the connection function copula to fit the marginal distribution function corresponding to each drought index. By integrating the marginal distribution function, the joint distribution functions corresponding to meteorological-hydrological, hydrological-agricultural and agricultural-groundwater droughts are constructed respectively to determine the risk probability of causing the next level of drought in the tested area under different meteorological drought levels.

[0024] Optionally, different types of drought can be divided into four levels: mild, moderate, severe, and extreme.

[0025] The data of SPEI and SRI corresponding to the propagation time in the above step S102 are used as input, and the marginal distribution function is obtained by adopting the non-parametric kernel density estimation method. The specific calculation formula is as follows: ; in, is a random sample from an unknown distribution, is the window width, n is the number of samples, x is the data point that needs to be estimated, i is the random sample sequence number.

[0026] Optionally, a multivariate distribution It can be expressed by the copula function, and the calculation formula is as follows: ; in, is a random variable x i The cumulative distribution function of X i is a random variable x i The corresponding drought state, ( u 1 , u 2 , …, u n ) is the probability value‌.

[0027] As shown in Table 1, Archimedean Copula functions Frank, Clayton and Gumbel, and elliptic Copula functions Gaussian and t are selected as candidates to construct the joint distribution function.

[0028] Table 1 copula function expression:

[0029] In Table 1, is the inverse function of the Gaussian distribution function, t k -1 The degrees of freedom are k The inverse function of the univariate t distribution function.

[0030] Furthermore, the Euclidean squared Euclidean distance was used to compare the differences between the empirical joint distribution and the theoretical joint distribution of the five copula functions. Based on the principle of minimizing the Euclidean squared Euclidean distance, the fit of the five copula functions was tested, and the smallest copula function type was selected.

[0031] Optionally, the fit of the copula function is tested using the minimum squared Euclidean distance principle. The specific calculation formula is as follows: ; in, is the empirical function value of the two-dimensional copula, is the theoretical function value of the two-dimensional copula, l Represents the size of the Euclidean square distance.

[0032] Optionally, based on the Bayesian probability model, the probability of inducing different levels of hydrological drought under different levels of meteorological drought is determined by analyzing the conditional probability of hydrological drought relative to different degrees of meteorological drought.

[0033] Optionally, the specific calculation formula for the probability of different levels of meteorological drought inducing different levels of hydrological drought is as follows: ; For example, the specific calculation formula for the probability of inducing mild hydrological drought under mild meteorological drought stress is as follows: ; The specific calculation formula for the probability of inducing mild hydrological drought under extreme meteorological drought stress is as follows: ; in, F SPEI is the cumulative distribution function corresponding to meteorological drought, F SRI is the cumulative distribution function corresponding to hydrological drought, SRI is the standardized runoff index corresponding to hydrological drought, SPEI It is the standardized precipitation evapotranspiration index corresponding to meteorological drought.

[0034] Optionally, by obtaining precipitation data and potential evapotranspiration PET data, the water surplus and deficit amount is obtained according to the precipitation data and the potential evapotranspiration PET data; then the water surplus and deficit amount is subjected to probability distribution fitting to obtain a fitted probability distribution, and the water surplus and deficit amount after the probability distribution fitting is standardized to obtain SPEI The value of .

[0035] For example, precipitation data and potential evapotranspiration (PET) data can be obtained through gridded precipitation and temperature data from the China Meteorological Data Service Center or through a dataset from the Climate Center of the University of East Anglia in the United Kingdom.

[0036] Optionally, obtain rainfall and runoff data, and obtain the edge probabilities corresponding to rainfall and runoff based on the rainfall and runoff data to determine SRI Value 。

[0037] Exemplarily, for example, rainfall and runoff data can be obtained from the monthly climate and climate water balance dataset of the global land surface provided by the University of Idaho.

[0038] S104: According to the risk probability of drought, iterate the meteorological drought index within a certain interval with a fixed sliding step length to obtain the drought risk probability results corresponding to each iteration. By presetting a drought risk probability value for the occurrence of hydrological drought, when the drought risk probability corresponding to each iteration is greater than or equal to the preset drought risk probability value, determine the interval where the corresponding meteorological drought index is located at this time as the trigger threshold for drought in the area to be measured. According to the change of the trigger threshold during the propagation process, obtain the drought resistance ability of the area to be measured.

[0039] Exemplarily, for a given level of hydrological drought, as the level of meteorological drought intensifies, the probability of inducing the occurrence of hydrological drought will theoretically approach 1.

[0040] Among them, when SRI is less than or equal to -0.5, it is considered that hydrological drought occurs.

[0041] Iterate SPEI from -0.5 at intervals of -0.1 each time until -3, so that the conditional probability corresponding to each iteration can be estimated. As Figure 3 shown, the first time is to calculate the probability that SRI ≤ -0.5 when -0.6 < SPEI ≤ -0.5 (because hydrological drought occurs when it is less than or equal to -0.5), the second time is to calculate the probability that SRI ≤ -0.5 when -0.7 < SPEI ≤ -0.6, the third time is to calculate the probability that SRI ≤ -0.5 when -0.8 < SPEI ≤ -0.7, and so on. The last time is to calculate the probability that SRI ≤ -0.5 when -3.0 < SPEI ≤ -2.9.

[0042] If the conditional probability is greater than or equal to 0.5, return the corresponding SPEI interval, and the right value of the interval will be regarded as the trigger threshold, that is, the SPEI corresponding to inducing hydrological drought at this level; if the conditional probability is less than 0.5, continue to iterate with a step length of 0.1; if the conditional probability of the last iteration is still less than 0.5, it is considered that the pixel has no trigger threshold.

[0043] It can be understood that when the trigger threshold is low, that is, the smaller the SPEI value, it means that a higher level of meteorological drought is required to cause the situation where SRI is less than or equal to -0.5, indicating that the hydrology has a strong ability to resist meteorological drought; on the contrary, when the trigger threshold is high, that is, the larger the SPEI value, it means that a lower level of meteorological drought will trigger the situation where SRI is less than or equal to -0.5, indicating that the runoff in this pixel has a weak ability to resist meteorological drought.

[0044] The above is a multi-type drought propagation process analysis method provided by one or more embodiments of this specification. Based on the same idea, this specification also provides a corresponding multi-type drought propagation process analysis device including: The acquisition module is used to obtain meteorological, hydrological, agricultural and groundwater drought data at different time scales in the test area; The correlation analysis module is used to determine the correlation between the meteorological drought index at different time scales and the hydrological drought index, agricultural drought index and groundwater drought index respectively according to the meteorological, hydrological, agricultural and groundwater drought data at different time scales, obtain the propagation time of meteorological-hydrological, meteorological-agricultural and meteorological-groundwater droughts, and determine the drought propagation path of meteorological-hydrological-agricultural-groundwater droughts; A construction module is used to construct drought propagation paths based on meteorological-hydrological-agricultural-groundwater drought. The nonparametric kernel density estimation method is combined with the connection function copula to fit the marginal distribution functions corresponding to each drought index. By integrating the marginal distribution functions, the joint distribution functions corresponding to meteorological-hydrological, hydrological-agricultural and agricultural-groundwater droughts are constructed respectively to determine the risk probability of causing the next level of drought in the tested area under different meteorological drought levels. The determination module is used to iterate the meteorological drought index within a certain interval with a fixed sliding step size according to the risk probability of drought, and obtain the drought risk probability result corresponding to each iteration. By presetting a drought risk probability value of hydrological drought, when the drought risk probability corresponding to each iteration is greater than or equal to the pre-set drought risk probability value, the interval of the meteorological drought index corresponding to this time is determined as the trigger threshold of drought in the test area, and the resistance of the test area to drought is obtained according to the change of the trigger threshold during the propagation process.

[0045] The present invention also provides a computer-readable storage medium, which stores a computer program. The computer program can be used to execute the multi-type drought propagation process analysis method provided above.

[0046] The present invention also provides Figure 4 The structural diagram of the computer device shown in FIG. Figure 4 As shown, at the hardware level, the computer device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory, and may also include hardware required for other services. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to implement the multi-type drought propagation process analysis method provided in the above embodiment.

[0047] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of the present invention.

Claims

1. A multi-type drought propagation process analysis method, characterized in that: include: Obtain meteorological, hydrological, agricultural and groundwater drought data at different time scales in the test area; Based on the meteorological, hydrological, agricultural and groundwater drought data at different time scales, the correlation between the meteorological drought index at different time scales and the hydrological drought index, agricultural drought index and groundwater drought index was determined, the propagation time of meteorological-hydrological, meteorological-agricultural and meteorological-groundwater droughts was obtained, and the drought propagation path of meteorological-hydrological-agricultural-groundwater drought was determined; Based on the drought propagation path of meteorological-hydrological-agricultural-groundwater drought, the non-parametric kernel density estimation method is used in combination with the connection function copula to fit the marginal distribution function corresponding to each drought index. By integrating the marginal distribution function, the joint distribution functions corresponding to meteorological-hydrological, hydrological-agricultural and agricultural-groundwater droughts are constructed respectively to determine the risk probability of causing the second-level drought in the tested area under different meteorological drought levels. According to the risk probability of drought, the meteorological drought index is iterated within a certain interval with a fixed sliding step size to obtain the drought risk probability result corresponding to each iteration. By presetting a drought risk probability value for hydrological drought, when the drought risk probability corresponding to each iteration is greater than or equal to the pre-set drought risk probability value, the interval of the meteorological drought index corresponding to this time is determined as the trigger threshold of drought in the test area. According to the change of the trigger threshold during the propagation process, the resistance of the test area to drought is obtained.

2. The method for analyzing the multi-type drought propagation process according to claim 1, characterized in that: The correlation between the meteorological drought index and the hydrological drought index, the agricultural drought index and the groundwater drought index at different time scales is determined, specifically including: The Spearman correlation coefficient method was used to determine the maximum positive correlation between the meteorological drought index at different scales and the hydrological drought index, agricultural drought index and groundwater drought index.

3. The method for analyzing the multi-type drought propagation process according to claim 1, characterized in that: The method of using non-parametric kernel density estimation combined with the connection function copula to fit the marginal distribution function corresponding to each drought index specifically includes: The marginal distribution function is obtained using the following formula: ， in, is a random sample from an unknown distribution, is the window width, n is the number of samples, x is the data point that needs to be estimated, i is the random sample sequence number.

4. The method for analyzing the multi-type drought propagation process according to claim 1, characterized in that: The bivariate joint distribution function is constructed to determine the risk probability of causing a second-level drought in the tested area under different meteorological drought levels, specifically including: According to the bivariate joint distribution function and the Bayesian probability model, the conditional probability of hydrological drought relative to different levels of meteorological drought is analyzed to determine the risk probability of the next level of drought in the tested area under different levels of meteorological drought.

5. The method for analyzing the multi-type drought propagation process according to claim 1, characterized in that: The meteorological drought levels specifically include: Mild drought level, moderate drought level, severe drought level and extreme drought level.

6. The method for analyzing the multi-type drought propagation process according to claim 1, characterized in that: The fixed sliding step size is 0.

1.

7. A multi-type drought propagation process analysis device, characterized in that: include: The acquisition module is used to obtain meteorological, hydrological, agricultural and groundwater drought data at different time scales in the test area; The correlation analysis module is used to determine the correlation between the meteorological drought index at different time scales and the hydrological drought index, agricultural drought index and groundwater drought index respectively according to the meteorological, hydrological, agricultural and groundwater drought data at different time scales, obtain the propagation time of meteorological-hydrological, meteorological-agricultural and meteorological-groundwater droughts, and determine the drought propagation path of meteorological-hydrological-agricultural-groundwater droughts; A construction module is used to construct drought propagation paths based on meteorological-hydrological-agricultural-groundwater drought. The nonparametric kernel density estimation method is combined with the connection function copula to fit the marginal distribution functions corresponding to each drought index. By integrating the marginal distribution functions, the joint distribution functions corresponding to meteorological-hydrological, hydrological-agricultural and agricultural-groundwater droughts are constructed respectively to determine the risk probability of causing the next level of drought in the tested area under different meteorological drought levels. The determination module is used to iterate the meteorological drought index within a certain interval with a fixed sliding step size according to the risk probability of drought, and obtain the drought risk probability result corresponding to each iteration. By presetting a drought risk probability value of hydrological drought, when the drought risk probability corresponding to each iteration is greater than or equal to the pre-set drought risk probability value, the interval of the meteorological drought index corresponding to this time is determined as the trigger threshold of drought in the test area, and the resistance of the test area to drought is obtained according to the change of the trigger threshold during the propagation process.

8. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method for analyzing the multi-type drought propagation process according to any one of claims 1 to 6 is implemented.

9. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for analyzing the multi-type drought propagation process as claimed in any one of claims 1 to 6 is implemented.

Images