Soil water sudden drought prediction method and system

Abstract

The invention discloses a soil water sudden drought prediction method and system, and the method comprises the steps: obtaining the historical index data of each grid point, and obtaining the corresponding soil water drought condition data of each grid point at different moments according to a time step delta T; according to the soil water drought condition data corresponding to each lattice point at different moments, identifying soil water drought events through time history connection, and obtaining time-varying data of areas covered by each soil water drought event in the research area; constructing a random forest model for predicting the drought formation time F by taking the average rainfall P, the average temperature T and the distance value G as input quantities and the drought formation time F as an output quantity according to time-varying data of the areas covered by the soil water drought events; according to the soil water drought condition data corresponding to each grid point at different moments, performing statistics to obtain drought expansion rate data of each grid point; and predicting drought area development trend data. According to the invention, the accuracy of predicting the sudden drought event of soil water can be improved.

Description

Technical Field

[0001] The invention relates to a method for predicting drought events, and in particular to a method and system for predicting sudden drought of soil water. Background Art

[0002] Global warming has caused a sharp rise in global temperatures, and the energy balance and material exchange process between various spheres have changed significantly, bringing irreversible impacts to the climate system. Drought events are greatly affected by rising temperatures, and their evolution speed and intensity gradually increase. The more serious ones will form extreme sudden droughts. Extreme sudden droughts have the characteristics of greater losses and lower prediction accuracy than ordinary drought events.

[0003] How to predict extreme sudden drought is an important issue facing researchers today. Many studies analyze and evaluate the entire drought event by analyzing the characteristics of drought at a single grid point, ignoring the spatiotemporal characteristics of drought as a three-dimensional phenomenon. In addition, due to the suddenness of sudden drought events, how to accurately predict sudden drought events is also a problem that needs to be further solved in related research. Summary of the invention

[0004] Purpose of the invention: In view of the above problems, the present invention proposes a soil water sudden drought prediction method and system, which can improve the accuracy of soil water sudden drought event prediction and improve the early warning level of regional sudden drought.

[0005] Technical solution: The technical solution adopted by the present invention is a soil water sudden drought prediction method, comprising:

[0006] The study area is gridded, and the soil water drought data corresponding to each grid point at different times are obtained according to the time step ΔT based on the historical index data of each grid point;

[0007] According to the soil water drought data corresponding to each grid point at different times, soil water drought events are identified through time-course connection to obtain data on the changes in the area covered by each soil water drought event in the study area over time; according to the data on the changes in the area covered by each soil water drought event over time, with average precipitation P, average temperature T, and distance value G as inputs and drought formation time F as output, a random forest model for predicting drought formation time F is constructed; the distance value G is the minimum distance between the grid point and the edge of the drought patch;

[0008] According to the soil water drought condition data corresponding to each grid point at different times, the drought expansion rate data of each grid point is obtained by statistics;

[0009] Predict the development trend of drought area, including: setting G maxis the maximum value of the distance value G among the identified soil water drought events; for any grid point x, if G≤G max , according to the drought expansion rate data of each grid point, the expansion rate η of the grid point x is obtained; according to the average precipitation P, average temperature T, and distance value G of the grid point x, the drought formation time F of the grid point x is output through the random forest model; all grid points with F≤ΔT are screened out, and the grid points with expansion rate η greater than the probability threshold are determined to have drought at the next moment; the predicted time T is divided into several segments according to the step length ΔT, and the calculation of this step is repeated until the prediction ends at time T, so as to obtain the drought area development trend data;

[0010] Based on the predicted drought area development trend data, the normalized curve of cumulative drought area-time is calculated, and the suddenness of drought is judged based on the normalized curve of cumulative drought area-time.

[0011] The index data uses soil moisture SM.

[0012] Soil water humidity SM is used to judge the soil water drought. By calculating the percentile value of soil water humidity SM, it is judged as soil water drought if the percentile value of soil water humidity SM is lower than the set threshold.

[0013] Using soil moisture SM to judge soil water drought conditions includes the following processes: extracting daily soil moisture data within a selected period of time in the study area to obtain the average soil moisture data over a period of time, and then calculating the marginal distribution function of the average soil moisture data; using the KS test method and the root mean square error method to fit the optimal probability distribution function; according to the optimal probability distribution function, converting the soil moisture value into a quantile, judging that the soil water moisture SM quantile is lower than the set threshold is soil water drought.

[0014] The time course connection includes the following process: determining the coverage area of ​​drought events in two adjacent time periods, and when the overlap of the coverage areas of two drought events in adjacent time periods is higher than a setting, determining that they are the same drought event.

[0015] The ordinate of the drought cumulative area-time normalized curve is the normalized drought coverage area, and the abscissa is the normalized time. Judging the suddenness of drought according to the drought cumulative area-time normalized curve includes: calculating the slope of the drought cumulative area-time normalized curve, judging whether the slope is greater than a threshold value, and if so, predicting that the drought type is sudden drought.

[0016] The normalized drought coverage area is calculated as follows:

[0017]

[0018] Where N Ais the normalized drought coverage area; n is the number of divided time periods, A j is the maximum drought coverage area projected in period j; It is the cumulative sum of the maximum drought coverage area projected in each time period;

[0019] The normalized time is calculated as follows:

[0020]

[0021] Where N T is the normalized time; T j is the jth period of drought; It is the sum of drought events in different time periods.

[0022] The present invention proposes a soil water sudden drought prediction system, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the soil water sudden drought prediction method is implemented when the processor executes the computer program.

[0023] The present invention provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the soil water sudden drought prediction method is implemented.

[0024] The present invention also provides a computer program product, comprising a computer program / instruction, which implements the soil water sudden drought prediction method when executed by a processor.

[0025] Beneficial effects: Compared with the prior art, the present invention has the following advantages: the present invention identifies the changes in the area covered by drought events over time through time-course connection, and constructs a random forest model to predict the time of drought occurrence; by introducing the expansion rate, the development trend of drought is predicted, and the global drought situation is judged in combination with the prediction results of the random forest model. Compared with the existing drought prediction algorithm, the present invention can synchronously predict global drought data, greatly improve the accuracy of the suddenness of drought development, and can intuitively predict the trend of future soil water drought and judge the type of drought. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a flow chart of the soil water sudden drought prediction method of the present invention;

[0027] Figure 2 It is a schematic diagram of the drought area development trend data predicted according to the present invention. DETAILED DESCRIPTION

[0028] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0029] The soil water sudden drought prediction method of the present invention is as follows: Figure 1 As shown, the following steps are included. Follow the process described in the steps. There is no specific order between the steps:

[0030] Step 1: Grid the study area, obtain the index data corresponding to each grid point over a period of time, determine whether it is soil water drought based on the index data, and obtain the soil water drought data corresponding to each grid point at different times according to the time step ΔT (two weeks).

[0031] In this embodiment, soil moisture content SM is used to judge soil water drought. The percentile value of soil moisture content SM is calculated, and soil moisture content SM is judged to be soil water drought when it is lower than a set threshold value of 0.3.

[0032] Another method of using soil moisture SM to judge soil water drought is to extract daily soil moisture data within a selected time in the study area to obtain the average soil moisture data within two weeks. Then calculate its marginal distribution function, and the function type can select exponential distribution, lognormal distribution, generalized extreme value distribution, etc. The KS test method and the root mean square error method (RMSE) are used to fit the best probability distribution function. After obtaining the probability distribution function, the soil moisture value can be converted into a percentile value to facilitate the identification of sudden drought. In this embodiment, the grid points with a percentile value lower than 0.3 are judged as drought. For different regions, the critical value should change with local standards.

[0033] Step 2: Based on the soil water drought data corresponding to each grid point at different times, soil water drought events are identified through time-course connection to obtain data on the temporal changes in the area covered by each soil water drought event in the study area.

[0034] The time-course connection includes the following processes: (1) In the same time period (such as within a week), for a grid point identified as drought, that is, a grid point with a percentile value lower than 0.3, if there are grid points identified as drought in the surrounding 3×3 area, these grid points are considered to be covered by the same drought event until there are no new drought grid points within the 3×3 range of the edge grid points of the drought event. After filtering out drought events with too small coverage areas, the different drought events that occurred in the study area during this period and their covered areas can be divided. (2) In the next period, the same operation is performed to obtain the different drought events that occurred in the next period and their covered areas. (3) The drought coverage areas of the two periods are determined. It is believed that when the overlap of the coverage areas of two droughts in consecutive periods is higher than a certain value, the two droughts can be determined as the expansion or contraction of the same drought event after time advances. Minimum drought coverage area A min With minimum overlap A oThe selection of these two parameters is very important. They may have different values ​​in different regions, so parameter calibration is carried out. Different combinations of minimum drought coverage area and overlap parameters are set, and the obtained drought event evolution process is fitted with the actual drought event evolution process in the study area. The parameter combination with the best fitting effect is selected as the parameter combination of the study area.

[0035] Step 3: Based on the soil water drought data corresponding to each grid point at different times, the drought expansion rate data of each grid point is statistically obtained.

[0036] The drought expansion rate data refers to the probability of drought occurring at the center grid point in a 3×3 area at a certain time after ΔT, when the surrounding 8 grid points are drought or not. ΔT is the set time step. That is, when there are 0 to 8 drought grid points around at the same time, the probability of drought occurring at the center grid point in these 9 situations. The following example shows the calculation method of drought expansion rate when there are 0, 1, and 2 drought grid points around.

[0037] When there are 0 grid points of drought around:

[0038]

[0039] N0 represents the number of times drought occurred at the central grid point and the eight surrounding grid points did not experience drought within ΔT (two weeks), or drought occurred at the same time as the central grid point; N represents the total number of times drought occurred at the central grid point.

[0040] When there is one grid point drought i nearby:

[0041]

[0042] M i represents the number of droughts that occurred at grid point i in the total time series; N i Represents the number of times the central grid changes from non-drought to drought within ΔT (two weeks) after drought occurs at drought grid i (excluding the number of times drought occurs at two grids at the same time).

[0043] When there are two grid points i,j in the vicinity that are dry:

[0044]

[0045] M j represents the number of droughts that occurred at grid point j in the total time series; N j represents the number of times the central grid point changes from non-drought to drought within ΔT (two weeks) after drought occurs at drought grid point j (excluding the number of times drought occurs at two grid points at the same time); A ij represents the number of times drought occurs at grid points i and j within two weeks of ΔT, and drought occurs at two grid points after the center grid point; B ijRepresents the number of times drought occurs at grid points i and j within two weeks ΔT.

[0046] Step 4: Based on the data of the changes in the area covered by each soil water drought event over time, the random forest model is introduced to predict the drought formation time F.

[0047] Obtain the average precipitation and temperature of all grid points that have experienced drought outside the drought patch S in the identified three-dimensional drought event, take the average precipitation P, average temperature T, and distance value G as independent variables, and the drought formation time F as the dependent variable. Import the random forest model, randomly select 70% of the samples as the training set and 30% of the samples as the test set, and you can get the prediction model for the drought formation time. The distance value G is the minimum distance between the grid point and the edge of the drought patch S. The average precipitation P and average temperature T refer to the average precipitation and average temperature within ΔT of each grid point. The drought formation time F is the time required for the grid point to become drought from the beginning of the drought patch S of the event when the grid point is in an identified three-dimensional drought event.

[0048] It should be noted that if a grid point is a grid point in a drought patch, it is generally not included in the training set and test set. If the grid point (grid point in the drought patch) becomes drought-free after the drought event area changes, and enters drought again before the end of the drought event, the drought formation time of the grid point is the time when it changes from drought-free to drought; if there is a grid point that repeatedly changes from drought-free to drought-free and then to drought-free in a drought event, the drought formation time is the minimum value of the time required from the formation of the drought patch to its own change to drought and the time required from drought-free to drought.

[0049] Step 5: Predict drought area development trend data.

[0050] If a drought patch appears in the study area, when its area exceeds the set minimum drought coverage area A min When the area is large, its development trend can be predicted.

[0051] Setting G max , is the maximum distance value G among the identified historical drought events. Any grid point G>G max When , the grid point is not included in the next calculation. The expansion rate η of each grid point is used together with the formation time F to determine whether drought will occur at the grid point at the next moment. The prediction time T is divided into several segments according to the step length ΔT (two weeks), and the probability threshold is set to 60%. All grid points with F≤ΔT are screened out, and grid points with expansion rate η greater than or equal to 60% are determined to have drought at the next moment. Repeat the calculation of this step until the prediction ends at time T. Figure 2 As shown, it is the forecast of regional drought in a certain ΔT period. The forecast area of ​​drought in this period is the combination of green and red parts, which is also the initial area for calculation in the next period.

[0052] Step 6: Based on the predicted drought area development trend data, calculate the normalized curve of cumulative drought area-time to determine the suddenness of the drought.

[0053] The area where drought may develop under the selected time, so by continuously selecting time, the drought area of ​​1, 2, ..., n weeks can be obtained, that is, the development of drought area over time. The normalized curve is used to describe the development speed of drought under the threshold η, and different possible drought development situations can be obtained by changing the threshold η. The drought cumulative area-time normalized curve is used to describe the speed of change of the drought cumulative area over time, and the calculation formulas of the vertical and horizontal coordinate values ​​are:

[0054]

[0055] In the formula, the total duration of drought is p periods, and j is the current drought period. A is the normalized drought coverage area; A i is the maximum drought projection area in period i; is the cumulative maximum drought projection area for each period, It is the cumulative maximum drought projection area up to period j.

[0056]

[0057] Where N T is the normalized time; T j is the jth period of drought; It is the sum of drought events in different time periods.

[0058] N A As the ordinate of the curve, N T As the horizontal axis of the curve, the normalized curve of the cumulative area-time of a drought event is obtained. It is believed that when the curve is a slanted line with a slope of 1, it is a special state in which the area increases uniformly over time, that is, the drought type is not a sudden drought. In general, it is believed that if there is a period of time when the slope of the curve is greater than 1, the drought in that period can be judged as a sudden drought. The greater the slope, the greater the suddenness of the drought.

[0059] In one embodiment, a soil water sudden drought prediction system is provided, 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 soil water sudden drought prediction method when executing the computer program.

[0060] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned soil water sudden drought prediction method is implemented.

[0061] In one embodiment, a computer program product is provided, comprising a computer program / instruction, which implements the soil water sudden drought prediction method when executed by a processor.

[0062] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0063] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0064] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0065] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

Claims

1. A soil water sudden drought prediction method, characterized in that: include: The study area is gridded, and the soil water drought data corresponding to each grid point at different times are obtained according to the time step ΔT based on the historical index data of each grid point; According to the soil water drought data corresponding to each grid point at different times, soil water drought events are identified through time-course connection to obtain data on the changes in the area covered by each soil water drought event in the study area over time; According to the data of the area covered by each soil water drought event changing over time, with average precipitation P, average temperature T, and distance value G as input, and drought formation time F as output, a random forest model for predicting drought formation time F is constructed; the distance value G is the minimum distance between the grid point and the edge of the drought patch; According to the soil water drought condition data corresponding to each grid point at different times, the drought expansion rate data of each grid point is obtained by statistics; Predict the development trend of drought area, including: setting G max is the maximum value of the distance value G among the identified soil water drought events; For any grid point x, if G≤G max , according to the drought expansion rate data of each grid point, the expansion rate η of the grid point x is obtained; according to the average precipitation P, average temperature T, and distance value G of the grid point x, the drought formation time F of the grid point x is output through the random forest model; all grid points with F≤ΔT are screened out, and the grid points with expansion rate η greater than the probability threshold are determined to have drought at the next moment; the predicted time T is divided into several segments according to the step length ΔT, and the calculation of this step is repeated until the prediction ends at time T, so as to obtain the drought area development trend data; Based on the predicted drought area development trend data, the normalized curve of cumulative drought area-time is calculated, and the suddenness of drought is judged based on the normalized curve of cumulative drought area-time.

2. The soil water sudden drought prediction method according to claim 1, characterized in that: The index data uses soil moisture SM.

3. The soil water sudden drought prediction method according to claim 2, characterized in that: Soil water humidity SM is used to judge the soil water drought. By calculating the percentile value of soil water humidity SM, it is judged as soil water drought if the percentile value of soil water humidity SM is lower than the set threshold.

4. The soil water sudden drought prediction method according to claim 2, characterized in that: Using soil moisture SM to judge soil water drought conditions includes the following processes: extracting daily soil moisture data within a selected period of time in the study area to obtain the average soil moisture data over a period of time, and then calculating the marginal distribution function of the average soil moisture data; using the KS test method and the root mean square error method to fit the optimal probability distribution function; according to the optimal probability distribution function, converting the soil moisture value into a quantile, judging that the soil water moisture SM quantile is lower than the set threshold is soil water drought.

5. The soil water sudden drought prediction method according to claim 1, characterized in that: The time course connection includes the following process: determining the coverage area of ​​drought events in two adjacent time periods, and when the overlap of the coverage areas of two drought events in adjacent time periods is higher than a setting, determining that they are the same drought event.

6. The soil water sudden drought prediction method according to claim 1, characterized in that: The ordinate of the drought cumulative area-time normalized curve is the normalized drought coverage area, and the abscissa is the normalized time. Judging the suddenness of drought according to the drought cumulative area-time normalized curve includes: calculating the slope of the drought cumulative area-time normalized curve, judging whether the slope is greater than a threshold value, and if so, predicting that the drought type is sudden drought.

7. The soil water sudden drought prediction method according to claim 1, characterized in that: The normalized drought coverage area is calculated as follows: Where N A is the normalized drought coverage area; n is the number of divided time periods, A j is the maximum drought coverage area projected in period j; It is the cumulative sum of the maximum drought coverage area projected in each time period; The normalized time is calculated as follows: Where N T is the normalized time; T h is the jth period of drought; It is the sum of drought events in each period.

8. A soil water sudden drought prediction system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the soil water sudden drought prediction method according to any one of claims 1 to 7 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the soil water sudden drought prediction method according to any one of claims 1 to 7 is implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the soil water sudden drought prediction method according to any one of claims 1 to 7 is implemented.

Images