Method and system for determining mutual feedback relationship between droughts

Abstract

The present invention discloses a method and system for determining the mutual feedback relationship of droughts. The method includes obtaining the drought index of the first drought and the drought index of the second drought; determining the joint distribution between the two droughts according to the two drought indices; determining the conditional probability P1 that the first drought occurs in the current time period under the condition that the first drought occurred in the previous time period and multiple conditional occurrence probabilities P2 that change with the second drought under the condition that the first drought occurred in the previous time period according to the joint distribution; and determining the mutual feedback relationship between the first drought and the second drought according to P1 and multiple P2s. The present invention can diagnose the mutual feedback direction between multiple influencing factors of droughts, calculate the negative-positive feedback conversion thresholds between droughts under different state scenarios, and quantify the feedback sensitivity of positive and negative feedbacks between droughts, which is helpful for the precise regulation of agricultural droughts.

Description

Technical Field

[0001] The invention discloses a method and system for determining a mutual feedback relationship between droughts, in particular a method and system for determining a mutual feedback relationship between droughts based on the state perspective of vine Copula, belonging to the technical field of mutual feedback between droughts. Background Art

[0002] Drought is one of the most serious natural disasters. It can cause water shortages, ecosystem destruction, agricultural production losses and other hazards, causing huge losses to the entire ecosystem and social development. At present, drought has attracted the attention of many researchers. According to the causes and impacts of drought, it can be divided into meteorological drought, hydrological drought, agricultural drought, ecological drought and socioeconomic drought. There are complex connections between different droughts. The formation and disaster-causing processes of drought often involve mutual feedback and interaction of multiple types, stages and scenarios. Over time, one drought will gradually lead to another drought, and the transmission of water loss between different types of drought is called drought propagation. Existing research focuses on the connection between different droughts or drought propagation. However, there is currently a lack of feedback research between droughts.

[0003] Taking meteorological drought and agricultural drought as examples, meteorological drought is defined as insufficient precipitation, and agricultural drought is defined as insufficient soil moisture. In the land-atmosphere coupled system, the feedback process between meteorological drought and agricultural drought is very complex. Many studies have shown that meteorological drought can be transmitted to agricultural drought to a certain extent. However, the lack of research on the feedback of agricultural drought to meteorological drought has led to poor drought early warning capabilities. Summary of the invention

[0004] The purpose of this application is to provide a method and system for determining the mutual feedback relationship between droughts, so as to solve the technical problem of poor drought early warning capability caused by the lack of feedback research between droughts in the prior art.

[0005] A first aspect of the present invention provides a method for determining a drought inter-feedback relationship, comprising:

[0006] Obtain the drought index of the first drought and the drought index of the second drought;

[0007] determining a joint distribution between two droughts based on the two drought indices;

[0008] According to the joint distribution, determine a conditional probability P1 of the first drought occurring in the current time period under the condition that the first drought occurred in the previous time period and a plurality of conditional probabilities P2 of the first drought occurring in the previous time period and changing with the second drought;

[0009] According to the P1 and the plurality of P2s, a mutual feedback relationship between the first drought and the second drought is determined.

[0010] Preferably, based on the P1 and multiple P2s, determine the mutual feedback relationship between the first type of drought and the second type of drought, which specifically includes:

[0011] When P2 < P1, then the second type of drought has a negative feedback on the first type of drought;

[0012] When P2 > P1, then the second type of drought has a positive feedback on the first type of drought.

[0013] Preferably, after determining the mutual feedback relationship between the first type of drought and the second type of drought, it further includes:

[0014] Based on the P1 and the P2, determine the conversion threshold for the mutual feedback between the first type of drought and the second type of drought.

[0015] Preferably, after determining the conversion threshold for the mutual feedback between the first type of drought and the second type of drought, it further includes:

[0016] Based on the conversion threshold, determine the feedback sensitivity of the second type of drought to the first type of drought.

[0017] Preferably, based on the conversion threshold, determine the feedback sensitivity of the second type of drought to the first type of drought, which specifically includes:

[0018] Obtain the drought indices of multiple second types of droughts less than the conversion threshold and the corresponding P2s for each drought index, and in combination with the method of curve fitting, determine the positive feedback sensitivity of the second type of drought to the first type of drought;

[0019] Obtain the drought indices of multiple second types of droughts greater than or equal to the conversion threshold and the corresponding P2s for each drought index, and in combination with the method of curve fitting, determine the negative feedback sensitivity of the second type of drought to the first type of drought.

[0020] Preferably, based on the two drought indices, determine the joint distribution between the two types of droughts, which specifically includes:

[0021] Respectively obtain the marginal distributions of the drought index of the first type of drought in the previous time period, the drought index of the first type of drought in the current time period, and the drought index of the second type of drought in the current time period;

[0022] Using the vine copula function, obtain multiple preliminary joint distributions between the above multiple marginal distributions;

[0023] Select an optimal preliminary joint distribution from the multiple preliminary joint distributions as the joint distribution between the two types of droughts.

[0024] Preferably, the plurality of preliminary joint distributions include a normal copula joint distribution, a t copula joint distribution, a Gumble copula joint distribution, a Clayton copula joint distribution and a Frank copula joint distribution.

[0025] Preferably, selecting an optimal preliminary joint distribution from the plurality of preliminary joint distributions as the joint distribution between the two droughts specifically includes:

[0026] Combined with the Akaike information criterion, an optimal preliminary joint distribution is selected from the multiple preliminary joint distributions as the joint distribution between the two droughts.

[0027] Preferably, the drought index of the first type of drought and the drought index of the second type of drought are drought indices at the same time scale and the same spatial resolution.

[0028] A second aspect of the present invention provides a system for determining a mutual feedback relationship between droughts, comprising:

[0029] An index acquisition module, the index acquisition module is used to acquire a drought index of a first type of drought and a drought index of a second type of drought;

[0030] A distribution determination module, the distribution determination module is used to determine the joint distribution between two droughts according to the two drought indices;

[0031] A probability determination module, the probability determination module is used to determine, based on the joint distribution, a conditional probability P1 of the first drought occurring in the current time period under the condition that the first drought occurred in the previous time period and a plurality of conditional probabilities P2 of the first drought occurring in the previous time period and changing with the second drought;

[0032] A relationship determination module, wherein the relationship determination module is used to determine the mutual feedback relationship between the first type of drought and the second type of drought according to the P1 and the plurality of P2s.

[0033] Compared with the prior art, the method and system for determining the mutual feedback relationship between droughts of the present invention have the following beneficial effects:

[0034] The present invention constructs a method for determining the mutual feedback relationship between droughts from the state perspective of Copula vine, which can diagnose the mutual feedback direction between multiple influencing factors of drought, calculate the negative-positive feedback conversion threshold between droughts under different state scenarios, and quantify the feedback sensitivity of positive and negative feedback between droughts, which is helpful for precise regulation of agricultural drought and is important indicator information for irrigation. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1Schematic diagram of a flow chart of a method for determining a mutual feedback relationship between droughts in an embodiment of the present invention;

[0036] Figure 2 Schematic diagram of the process of step 1 in the embodiment of the present invention;

[0037] Figure 3 This is a schematic diagram of the process corresponding to steps 2 and 3 in an embodiment of the present invention;

[0038] Figure 4 It is a curve diagram of the change of P2 with the change of SSI in the embodiment of the present invention. DETAILED DESCRIPTION

[0039] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0040] like Figure 1 As shown, a first aspect of an embodiment of the present invention provides a method for determining a mutual feedback relationship between droughts, comprising:

[0041] Step 1: Obtain the drought index of the first drought and the drought index of the second drought, such as Figure 2 shown.

[0042] The present invention first collects gridded data such as precipitation, runoff, soil moisture, temperature, vegetation, etc. in the study area, and processes the data into the same time scale and the same spatial resolution to calculate different drought indices. The time scale can be monthly, ten-day, annual, etc.

[0043] The embodiment of the present invention takes the mutual feedback relationship between meteorological drought and agricultural drought as an example. Meteorological drought is the first type of drought, and agricultural drought is the second type of drought. Gridded data of precipitation and soil moisture at a monthly scale are collected, and the gridded scale data are processed into the same time scale and the same spatial resolution, and finally precipitation data and soil moisture data with consistent sequence lengths are obtained. The method for determining the mutual feedback relationship between droughts of the present invention can also be used to quantify the complex mutual feedback relationship between meteorological drought, hydrological drought, agricultural drought, and vegetation drought.

[0044] Then, the processed precipitation data and soil moisture data are used to calculate the drought index. The Standardized Precipitation Index (SPI) calculated from the precipitation data is used to characterize meteorological drought; the Standardized Soil Moisture Index (SSI) calculated from the soil moisture data is used to characterize agricultural drought. The embodiment of the present invention uses a drought index on a monthly scale.

[0045] The standardized precipitation index SPI is calculated, and the specific calculation process and formula are as follows:

[0046] First, the optimal distribution function of the precipitation sequence is selected, the cumulative probability is calculated according to the optimal distribution function, and then the SPI is obtained through standardization. The calculation principle of SSI is the same as that of SPI, which will not be repeated here. The formula used is as follows:

[0047]

[0048] Where H(x) is the cumulative probability of a certain time scale; constant c0 = 2.515517;

[0049] c1=0.802853; c2=0.010328; d1=1.432788; d2=0.189269; d3=0.001308;

[0050]

[0051] Step 2: Determine the joint distribution between the two droughts based on the two drought indices, including:

[0052] Step 21, respectively obtain the drought index of the first type of drought in the current period and the previous period and the marginal distribution of the drought index of the second type of drought in the current period.

[0053] In the embodiment of the present invention, meteorological drought and agricultural drought are taken as examples, and step 21 is specifically as follows: obtaining the standardized precipitation index (SPI) of the current time period respectively. t ), Standardized Precipitation Index (SPI) for the previous period t-1 ) and the Standardized Soil Moisture Index (SSI) for the current period t )’s marginal distribution.

[0054] Step 22: Use the vine copula function to obtain multiple preliminary joint distributions between the above multiple marginal distributions.

[0055] The embodiment of the present invention is based on the rattan copula theory and the Bayesian framework, according to the standardized precipitation index (SPI) of the current time period t), Standardized Precipitation Index (SPI) for the previous period t-1 ) and the Standardized Soil Moisture Index (SSI) for the current period t ) to construct the Standardized Precipitation Index (SPI) for the current time period t ), Standardized Precipitation Index (SPI) for the previous period t-1 ) and the Standardized Soil Moisture Index (SSI) for the current period t ) multiple preliminary joint distributions. The calculation principle of vine copula is simply to decompose the multidimensional copula into multiple two-dimensional copulas. R vine copula is used, and 5 copulas are used on each edge to perform optimal pairing of two-dimensional copulas. The 5 copulas are normal copula, t copula, Gumble copula, Clayton copula, and Frank copula. Among them, the two-dimensional copula selected when coupling the vine copula model can also be replaced by other copulas. Then, the optimal copula function is selected according to the AIC Akaike information criterion to connect the vine copula structure.

[0056] Step 23: Selecting an optimal preliminary joint distribution from the multiple preliminary joint distributions as the joint distribution between the two droughts, specifically comprising:

[0057] Combined with the Akaike information criterion (AIC), an optimal copula function was selected from multiple preliminary joint distributions to connect the rattan copula structure as the joint distribution between the two droughts.

[0058] Step 3: According to the joint distribution, determine the conditional probability P1 of the first drought occurring in the current time period under the condition that it occurred in the previous time period and the multiple conditional probabilities P2 of the first drought occurring in the previous time period with the change of the second drought, such as Figure 3 shown.

[0059] The preset drought level in the embodiment of the present invention may be light drought, moderate drought, severe drought, etc.

[0060] The present invention refers to the idea of ​​Granger causality analysis, adopts the mutual causality of interactive objects, constructs a multi-scenario drought feedback relationship research method, and estimates the conditional probability of multiple interactive combinations of multiple drought scenarios. For example, taking the mild drought state as an example, P1 is as formula (2), and P2 is as formula (3):

[0061]

[0062]

[0063] Where P1 and P2 are the conditional probabilities of meteorological drought occurring under different conditions, t is the current month, and t - 1 is the previous month. An SPI is established using the optimal copula function determined in step 2 t (meteorological drought in the current month), SPI t-1 (meteorological drought in the previous month), SSI t (agricultural drought in the current month) joint distribution and calculate the conditional probability. Where ssi is a gradually changing range value. Exemplarily, ssi ranges from 2 to -2, that is, the soil moisture drought changes from wet to dry. Therefore, the P2 value is constantly changing. In an embodiment of the present invention, with ssi as the x-axis and P2 as the y-axis, a curve showing the change of P2 with the change of ssi is drawn, as Figure 4 shown

[0064] Step 4. Determine the mutual feedback relationship between the first type of drought and the second type of drought according to P1 and multiple P2s, specifically including:

[0065] When P2 < P1, the second type of drought has a negative feedback on the first type of drought, that is, it means that when agricultural drought occurs, it inhibits the intensification of meteorological drought;

[0066] When P2 > P1, the second type of drought has a positive feedback on the first type of drought, that is, it means that when agricultural drought occurs, it promotes the intensification of meteorological drought.

[0067] The method of the present invention can diagnose the mutual feedback direction between multiple influencing factors of drought.

[0068] Further, after determining the mutual feedback relationship between the first type of drought and the second type of drought, it further includes:

[0069] Step 5. Determine the conversion threshold of the mutual feedback between the first type of drought and the second type of drought according to P1 and P2, specifically: when P2 = P1, the corresponding SSI value is the conversion threshold of the negative - positive feedback of agricultural drought on meteorological drought, as Figure 4 shown that the conversion threshold is -0.45.

[0070] In order to better understand the different feedback degrees of agricultural drought on meteorological drought under positive feedback and negative feedback, after determining the conversion threshold of the mutual feedback between the first type of drought and the second type of drought, it further includes:

[0071] Step 6. Determine the feedback sensitivity of the second type of drought to the first type of drought according to the conversion threshold, specifically including:

[0072] Obtain the drought indices of multiple second types of droughts less than the conversion threshold and the corresponding P2s for each drought index, and combine the curve fitting method to determine the positive feedback sensitivity of the second type of drought to the first type of drought;

[0073] The drought indices of multiple second droughts that are greater than or equal to the conversion threshold and the P2 corresponding to each drought index are obtained, and the negative feedback sensitivity of the second drought to the first drought is determined by combining the curve fitting method.

[0074] The curve fitting method used in the embodiment of the present invention is preferably the least square method, and the obtained curves are y1=a1x1+b1 and y2=a2x2+b2, where a1 is the negative feedback sensitivity and a2 is the positive feedback sensitivity. When the sensitivity value is greater than 0, it indicates that P2 increases as SSI changes from 2 to -2. The larger the absolute value of the sensitivity, the more sensitive P2 is to SSI, and vice versa.

[0075] The present invention constructs a method for determining the mutual feedback relationship between droughts from the state perspective of Copula vines, which can diagnose the mutual feedback direction between multiple influencing factors of drought, calculate the negative-positive feedback conversion threshold between droughts under different state scenarios, and quantify the feedback sensitivity of positive and negative feedback between droughts, which is helpful for precise regulation of agricultural drought, is important indicator information for irrigation, helps to strengthen drought early warning capabilities, and ensure ecological and environmental safety.

[0076] A second aspect of an embodiment of the present invention provides a system for determining a mutual feedback relationship between droughts, including an index acquisition module, a distribution determination module, a probability determination module and a relationship determination module.

[0077] The index acquisition module is used to obtain the drought index of the first type of drought and the drought index of the second type of drought;

[0078] The distribution determination module is used to determine the joint distribution between two droughts according to two drought indices;

[0079] The probability determination module is used to determine, based on the joint distribution, a conditional probability P1 of the first drought occurring in the current time period under the condition that the first drought occurred in the previous time period and a plurality of conditional probabilities P2 of the first drought occurring in the previous time period and changing with the second drought;

[0080] The relationship determination module is used to determine the mutual feedback relationship between the first drought and the second drought according to P1 and multiple P2.

[0081] The present invention studies the mutual feedback relationship between droughts. Specifically, the formation and disaster-causing process of drought is complex, often involving mutual feedback and interaction of multiple types, stages and scenarios. Some studies focus on the connection between different droughts or drought propagation, but lack feedback research between droughts. The present invention constructs a model for calculating the mutual feedback relationship between droughts, thereby quantifying the mutual feedback relationship between droughts.

[0082] The present invention calculates the negative-positive feedback conversion threshold of agricultural drought to meteorological drought. Specifically, previous studies on the drought threshold framework have focused on the drought propagation threshold, the drought-triggered vegetation loss threshold or the food loss threshold. There is currently a lack of quantitative methods for the negative-positive feedback conversion threshold of agricultural drought to meteorological drought. The present invention can calculate the negative-positive feedback conversion threshold of agricultural drought to meteorological drought and the feedback sensitivity of negative feedback and positive feedback.

[0083] The present invention uses rattan copula to construct a model. Simple linear relationship structures and common copula connection functions cannot well describe the complex relationship between droughts. The rattan copula structure is flexible and can be better used for multivariate connection. The present invention uses rattan copula structure to construct the joint distribution between droughts. The resulting distribution is more accurate, which helps to accurately control agricultural droughts and is an important indicator of irrigation.

[0084] The present invention proposes a quantitative method for the mutual feedback relationship between droughts based on rattan copula, and the conversion threshold and feedback sensitivity of the negative and positive feedback of agricultural drought to meteorological drought under the state scenario are calculated to help enhance the drought early warning capability and ensure the safety of the ecological environment.

[0085] In the context of global warming, the method and system of the present invention can accurately monitor and alleviate drought, enhance the understanding of the mutual feedback relationship between droughts, facilitate the adoption of corresponding early warning and management measures, prevent drought and reduce disasters, and ensure the sustainable development of the ecology.

[0086] The above are only a few embodiments of the present application and do not constitute any form of limitation to the present application. Although the present application is disclosed as above with preferred embodiments, it is not intended to limit the present application. Any technician familiar with the profession, without departing from the scope of the technical solution of the present application, using the technical content disclosed above to make slight changes or modifications are equivalent to equivalent implementation cases and fall within the scope of the technical solution.

Claims

1. A method for determining the mutual feedback relationship between droughts, characterized in that: include: Obtain the drought index of the first drought and the drought index of the second drought; determining a joint distribution between two droughts based on the two drought indices; According to the joint distribution, the conditional probability of the first type of drought occurring in the current time period under the condition that it occurred in the previous time period is determined. P 1 and the multiple conditional probabilities of the first drought occurring in the previous period with the change of the second drought P 2; According to the P 1 or more of the P 2. Determine the mutual feedback relationship between the first drought and the second drought, specifically including: when P 2< P When 1, the second drought has a negative feedback on the first drought; when P 2> P When 1, the second drought has a positive feedback on the first drought; After determining the mutual feedback relationship between the first drought and the second drought, the method further includes: According to the P 1 and as stated P 2. determining a transition threshold of mutual feedback between the first drought and the second drought; determining, according to the conversion threshold, a feedback sensitivity of the second type of drought to the first type of drought; According to the two drought indices, a joint distribution between the two droughts is determined, specifically comprising: Obtain the marginal distribution of the drought index of the first type of drought in the previous period, the drought index of the first type of drought in the current period, and the drought index of the second type of drought in the current period respectively; Using the vine copula function, multiple preliminary joint distributions between the above multiple marginal distributions are obtained; An optimal preliminary joint distribution is selected from the multiple preliminary joint distributions as the joint distribution between the two droughts.

2. The method for determining the mutual feedback relationship between droughts according to claim 1, characterized in that: Determining the feedback sensitivity of the second type of drought to the first type of drought according to the conversion threshold specifically includes: Obtain a plurality of drought indices of the second type of drought that are less than the conversion threshold and a corresponding drought index of each drought index P 2. Determine the positive feedback sensitivity of the second drought to the first drought by combining the curve fitting method; Obtain a plurality of drought indices of the second type of drought that are greater than or equal to the conversion threshold and a corresponding drought index of each drought index. P 2. Determine the negative feedback sensitivity of the second type of drought to the first type of drought by combining the curve fitting method.

3. The method for determining the mutual feedback relationship between droughts according to claim 1, characterized in that: The plurality of preliminary joint distributions include a normal copula joint distribution, a t copula joint distribution, a Gumble copula joint distribution, a Clayton copula joint distribution and a Frank copula joint distribution.

4. The method for determining the mutual feedback relationship between droughts according to claim 1, characterized in that: Selecting an optimal preliminary joint distribution from the multiple preliminary joint distributions as the joint distribution between the two droughts specifically includes: Combined with the Akaike information criterion, an optimal preliminary joint distribution is selected from the multiple preliminary joint distributions as the joint distribution between the two droughts.

5. The method for determining the mutual feedback relationship between droughts according to claim 1, characterized in that: The drought index of the first type of drought and the drought index of the second type of drought are drought indices at the same time scale and the same spatial resolution.

6. A system for determining a mutual feedback relationship between droughts based on the method for determining a mutual feedback relationship between droughts according to any one of claims 1 to 5, characterized in that: include: An index acquisition module, the index acquisition module is used to acquire a drought index of a first type of drought and a drought index of a second type of drought; A distribution determination module, the distribution determination module is used to determine the joint distribution between two droughts according to the two drought indices; A probability determination module, the probability determination module is used to determine the conditional probability of the first type of drought occurring in the current time period under the condition that it occurred in the previous time period according to the joint distribution. P 1 and the multiple conditional probabilities of the first drought occurring in the previous period with the change of the second drought P 2; A relationship determination module, the relationship determination module is used to determine the relationship according to the P 1 or more of the P 2. Determine the mutual feedback relationship between the first drought and the second drought.

Images