A method, storage medium and device for detecting mutation points of climate model index sequence

Abstract

The present invention relates to a method, storage medium and device for detecting mutation points of a climate mode index sequence. The observation sequence data is read, the cutoff length and probability level of the climate mode are determined, and the statistically significant differences between different modes are calculated. A loop is started to search for mutation points. If the value of a certain point is significantly different from that of the previous climate mode, the point is considered to be a candidate point for climate mode mutation, otherwise it is moved to the next point. After the candidate point is determined, the mode conversion index of the point is calculated. If the index is negative, it means that the mode mutation fails. If all are positive, it means that the mutation is successful, and the point is officially confirmed as a mode mutation point. The new mode is used as the reference mode, and the above operation is continued until the end of the observation sequence, and all mode mutation points are detected. The present invention solves the problem that the detection performance of the existing method drops sharply at the end of the sequence. The proposed method can detect the climate mode transfer as early as possible and then monitor the change of its amplitude over time. The method is simple to operate, and the user can change the input parameters according to the needs.

Description

Technical Field

[0001] The invention relates to a method for detecting mutation points of a climate model index sequence, and in particular to a process algorithm for calculating and identifying mutation points of a climate model. Background Art

[0002] The changes in different phases of climate patterns are of great value in studying climate change. Existing methods for detecting abrupt changes in climate patterns usually use confirmatory statistical techniques and a priori assumptions about the timing of changes. Although there are some methods that can automatically detect discontinuities in time series, the performance of existing methods drops sharply at the end of the series because sufficient data (at least 10 years or more) must be accumulated to apply formal statistical tests. In addition, when sufficient data are available, the climate system may be about to switch to the opposite state again. Summary of the invention

[0003] In view of the above-mentioned deficiencies in the prior art, the present invention provides a method, storage medium and device for detecting mutation points of a climate pattern index sequence. The method makes up for the deficiencies of the existing methods and can detect climate pattern mutations with minimal delay and subsequently monitor changes in its amplitude over time, regardless of whether the incoming data sequence is stored in the form of outliers or absolute values.

[0004] A method for detecting mutation points of a climate model index sequence is used to detect mutation points of a diachronic or real-time online climate model index sequence data and visualize and intuitively display the method, comprising the following steps:

[0005] 1) Read observation sequence data;

[0006] 2) Determine the cutoff length and probability level of climate models and calculate the statistically significant differences between different models;

[0007] 3) Looping to find the mutation point starting from the next data point of the pattern length;

[0008] 4) Calculate the mean value of the previous mode R1 and the value that should be achieved in the subsequent cutoff length year when transitioning to the new mode R2;

[0009] 5) If the value of a certain point is significantly different from the previous climate model, then the point is considered to be a candidate point for climate model mutation. Otherwise, move to the next data point, incorporate the data point into the previous climate model R1, and return to step 4);

[0010] 6) Calculate the mode transfer index of all variables within the cutoff length from the candidate point in turn. If the mode transfer index of all variables is positive, determine this candidate point as the mode mutation point, update the previous mode R1 to the new mode R2 after the mode mutation point, and return to step 4); otherwise, move to the next candidate point, incorporate the candidate point into the previous climate mode R1, and repeat steps 4)-6);

[0011] 7) Traverse all observation sequence data to end the loop processing and detect all pattern mutation points;

[0012] 8) Visualize and intuitively draw a graph of the pattern mutation points of the current observed sequence data.

[0013] The method of determining the cutoff length and probability level of the climate model and calculating the statistically significant difference between different models is as follows: defining the climate model index x model cutoff length l and probability level p, and calculating the difference value diff between the average values ​​of the two models before and after the model mutation point according to the Student's t test.

[0014] The step of cyclically searching for a mutation point starting from the next point of the pattern length is specifically as follows: performing detection from each value of the j=l+1th value.

[0015] The calculation of the previous mode mean is specifically as follows: calculating the average value of the first l values ​​of the j-th position of the variable X As an estimate of the previous climate model R1; the value that should be reached in the subsequent cut-off length year to transition to the new model is: Calculate the value that meets the transition to the new model R2 and reaches in the subsequent l value

[0016] The point is considered as a candidate point for climate pattern mutation. The specific judgment is: the value x of the climate pattern variable at the location j Exceed

[0017] The operation for judging the candidate point of climate mode mutation failure is: incorporating the data point into the previous mode and recalculating the estimate of the previous climate mode R1. And move backward to point j+1.

[0018] The mode shift index is RSI: the index represents the accumulation of normalized outliers;

[0019] If i>j, and the index i increases and all the mode transfer indices of all i positions are positive, it means that the mode mutation occurs at position i. Then this point is taken as the starting point of the new climate model R2, and its estimated value is calculated. And take R2 as the new reference mode, Move to point j+1 and continue detection according to the above steps.

[0020] If the mode transfer index is negative, it means that there is no mode mutation at position i. This point is included in the previous mode R1 and the average value of the previous climate mode R1 is recalculated. And move backward to point j+1.

[0021] A computer-readable storage medium for detecting mutation points of a climate pattern index sequence, wherein a computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, the method steps for mutation point detection as described above are implemented to realize mutation point detection of a diachronic or real-time online climate pattern index sequence.

[0022] A device for detecting mutation points of a climate pattern index sequence comprises: a memory and a processor. The memory stores a program, and the processor loads the program to execute the method steps for mutation point detection as described above, thereby realizing mutation point detection and visual and intuitive display of a diachronic or real-time online climate pattern index sequence.

[0023] The present invention has the following beneficial effects and advantages:

[0024] 1. The present invention provides a method for detecting mutation points of a climate pattern index sequence, including a complete set of mutation point detection processes from data reading, data processing, calculation, and result visualization.

[0025] 2. This algorithm is completed only using IDL program, which is simple to operate and easy to implement.

[0026] 3. This algorithm is highly flexible. Users can detect mutation points in any climate data sequence according to their needs and identify mutation points that meet any confidence level. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is a flow chart of the method of the present invention;

[0028] Figure 2 This is a visualization result diagram of the climate index mutation point detected by the example of the method of the present invention. DETAILED DESCRIPTION

[0029] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation method of the present invention is described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the invention, so the present invention is not limited by the specific implementation disclosed below.

[0030] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0031] The following implementation takes the Pacific Decadal Oscillation Index PDO mutation point detection from January 1900 to 2003 as an example to illustrate the purpose of the present invention, but is not intended to limit the scope of the present invention.

[0032] In addition to climate indices, this method can be applied to any data series such as large-scale circulation index, sea ice extent anomaly index, sea ice output flux index, etc. The mutation point detection is performed on the above data to obtain the turning point of the data pattern.

[0033] The present invention solves the problem that the detection performance of existing methods drops sharply at the end of the sequence. It is proposed to detect climate mode shifts as early as possible and then monitor the changes in their amplitude over time.

[0034] like Figure 1 As shown, the program language reads the PDO sequence data, sets the cutoff length and probability level of the climate mode, calculates the statistically significant difference values ​​between different modes, and starts from the next point of the mode length to cyclically search for mutation points to determine whether there is a significant difference between the value of a certain point and the previous climate mode. If so, the mode transfer coefficient is calculated, otherwise it moves to the next point. If the mode transfer index is positive, this point is determined as the mode mutation point, and the new mode is used as the benchmark. Otherwise, it moves to the next point and incorporates this point into the previous climate mode. Repeat the above steps to perform mode mutation point detection. It includes the following steps:

[0035] Step 1): After setting the climate model index variable X model cutoff length l years and the probability level p, calculate the difference between the average values ​​of the first and last two models according to the Student t test (this value is statistically significant). The calculation method is as follows:

[0036] Formula 1:

[0037] To test each value of the jth position = l+1, first calculate the average of the first l values ​​of the index variable X at the jth position. As an estimate of the previous climate model R1, calculate the value that should be achieved in the subsequent cut-off length year when transitioning to the new model R2 The calculation method is as follows:

[0038] Formula 2:

[0039] If the value of the climate mode index variable at the jth location is x j The mean of the previous climate model If there is a significant difference, that is, if Formula 3 or Formula 4 is satisfied, the point is considered to be a candidate point for climate pattern mutation.

[0040] Formula 3:

[0041] Formula 4:

[0042] Where t is the value of the t distribution with 2l degrees of freedom at a given probability level p. In this method, it is assumed that the variances of the two climate modes are the same and equal to the average variance of the variable time series with a sliding window of l.

[0043] When this location is detected as a candidate point for climate pattern mutation, the RSI of the pattern transition index of l points behind this point is calculated. The calculation method is as shown in Formula 4. For all i>j, if x j and If the differences have the same sign, the confidence that a mode mutation has indeed occurred is increased; if they have opposite signs, the opposite is true.

[0044] Formula 5:

[0045] Among them, when hour, when hour, Among them, if the current point i satisfies the transition to the new mode R2, then is the average value reached in the next l values.

[0046] If i>j, all i-position mode shift indexes RSI are positive, indicating that a sudden change in mode occurs at that position. This point is used as the starting point of the new climate model R2, and its estimated value is calculated. And R2 is used as the new reference mode (that is, ), move to point j = j + 1 and continue the test according to the above steps. This is to ensure that even if the actual duration of the R2 mode is less than 1 year, the time of the next mode mutation can be correctly determined. If the mode transfer index is negative, it means that there is no mode mutation at this location. The point is included in the previous mode and the estimate of the previous climate model R1 is recalculated. And move backward one point j=j+1.

[0047] Use the IDL for loop statement to iterate until all mutation points are detected. Figure 2The figure shows an intuitive display of the visual mapping. The pattern mutation points detected by the PDO index from 1900 to 2003 are located in 1911, 1922, 1945, 1958, 1977, 1989 and 2003. It can be seen that the climate pattern shifted at each of the above mutation points, which is of great significance for the study of climate change.

[0048] A method for detecting mutation points in a series of climate pattern indices is implemented using an IDL program.

[0049] Furthermore, the above method steps can be implemented by a program and stored on a computer-readable medium, and then connected to a memory to execute the program.

[0050] Furthermore, the computer-readable medium and the processor can be used as devices, and the processor loads a program to detect the mutation points of the diachronic or real-time online climate mode index series and visualize and intuitively display them. The above are the main features of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention, and these improvements and modifications should be considered as the protection scope of the present invention.

Claims

1. A method for detecting mutation points of a climate pattern index sequence, characterized in that: It is used to detect the mutation points of the diachronic or real-time online climate model index series data and visualize and display them intuitively, including the following steps: 1) Read observation sequence data; 2) Determine the cutoff length and probability level of climate models and calculate the statistically significant differences between different models; 3) Looping to find the mutation point starting from the next data point of the pattern length; 4) Calculate the mean of the previous climate model R1 and the value that should be reached in the subsequent cutoff length year when transitioning to the new model R2; the calculation of the mean of the previous model is as follows: calculate the average of the first l values ​​of the j-th position of the variable X As an estimate of the previous climate model R1; the value that should be reached in the subsequent cutoff length year to transition to the new model is: Calculate the value that meets the transition to the new model R2 and reaches in the subsequent l values 5) If the value of a certain point is significantly different from the previous climate model, then the point is considered to be a candidate point for climate model mutation. Otherwise, move to the next data point, incorporate the data point into the previous climate model R1, and return to step 4); the specific judgment of a point being considered as a candidate point for climate model mutation is: the value x of the climate model variable at the location j Exceed The operation for judging the candidate point of climate mode mutation failure is: incorporating the data point into the previous mode and recalculating the estimate of the previous climate mode R1. And move backward to point j+1; 6) Calculate the mode transfer index of all variables within the cutoff length from the candidate point in turn. If the mode transfer index of all variables is positive, determine this candidate point as the mode mutation point, update the previous mode R1 to the new mode R2 after the mode mutation point, and return to step 4); otherwise, move to the next candidate point, incorporate the candidate point into the previous climate mode R1, and repeat steps 4)-6); 7) Traverse all observation sequence data to end the loop processing and detect all pattern mutation points; 8) Visualize and intuitively draw a graph of the pattern mutation points of the current observed sequence data.

2. A method for detecting mutation points of a climate pattern index sequence according to claim 1, characterized in that: The method of determining the cutoff length and probability level of the climate model and calculating the statistically significant difference between different models is as follows: defining the climate model index x model cutoff length l and probability level p, and calculating the difference value diff between the average values ​​of the two models before and after the model mutation point according to the Student's t test.

3. A method for detecting mutation points of a climate pattern index sequence according to claim 2, characterized in that: The step of cyclically searching for a mutation point starting from the next point of the pattern length is specifically as follows: performing detection from each value of the j=l+1th value.

4. The method for detecting mutation points of a climate pattern index sequence according to claim 1, characterized in that: The mode shift index is RSI: the index represents the accumulation of normalized outliers; If i>j, and the index i increases and all the mode transfer indices of all i positions are positive, it means that the mode mutation occurs at position i. Then this point is taken as the starting point of the new climate model R2, and its estimated value is calculated. And take R2 as the new reference mode, Move to point j+1 and continue detection according to the above steps.

5. The method for detecting mutation points of a climate pattern index sequence according to claim 1, characterized in that: If the mode transfer index is negative, it means that there is no mode mutation at position i. This point is included in the previous mode R1 and the average value of the previous climate mode R1 is recalculated. And move backward to point j+1.

6. A computer-readable storage medium for detecting a mutation point in a climate pattern index sequence, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the method steps for mutation point detection as described in any one of claims 1 to 5, thereby realizing mutation point detection for a historical or real-time online climate pattern index sequence.

7. A device for detecting mutation points of a climate pattern index sequence, characterized in that: include: A memory and a processor, wherein a program is stored in the memory, and the processor loads the program to execute the method steps for mutation point detection as described in any one of claims 1 to 5, so as to realize mutation point detection and visual and intuitive display of a historical or real-time online climate pattern index sequence.

Images