A Method for Analyzing Factors Influencing Sea Surface Temperature in the Tropical Western Indian Ocean Based on Cross Wavelet

Abstract

This invention provides a method for analyzing the influencing factors of sea surface temperature (SST) in the tropical western Indian Ocean based on cross wavelets. It analyzes the interannual variation of the SST sequence over the past 47 years using methods such as moving average and moving t-test. Then, it decomposes the SST sequence into multiple fixed-frequency components using EEMD and analyzes the stationarity of the SST sequence through the monotonicity of the long-term trend term. Next, it performs FFT transform on the filtered SST sequence to analyze periodicity and analyzes the hidden multi-period nesting phenomenon in the frequency domain. Regarding the analysis of influencing factors, it first selects four parameters: precipitation, sea surface heat flux, total cloud cover, and longwave radiation in the tropical western Indian Ocean region. Then, it uses Morlet wavelets to analyze the periodic nesting phenomenon among these four parameters and determines the order of the main periods of each parameter's variation through wavelet variance. Finally, it uses cross wavelets to analyze the correlation relationships between precipitation and SST, sea surface heat flux and SST, total cloud cover and SST, and longwave radiation and SST. This method analyzes the SST variation process in the tropical western Indian Ocean from a combined time-frequency domain perspective, providing a more comprehensive analysis.

Description

Technical Field

[0001] This invention relates to the field of sea surface temperature influencing factor analysis using time-frequency domain methods, with the aim of developing a method for analyzing sea surface temperature influencing factors in the tropical western Indian Ocean based on cross wavelets. Background Technology

[0002] With global warming, ocean temperatures have become increasingly unstable, making the study of ocean temperature changes crucial for long-term sustainable development. While numerous reliable studies have been conducted by scholars both domestically and internationally on the factors influencing sea surface temperature (SST), research specifically on the tropical western Indian Ocean is scarce. Furthermore, given the periodic nature of SST changes, this method combines time and frequency domain perspectives. This invention will first analyze SST in the tropical western Indian Ocean using methods such as moving average, moving t-test, and MK mutation in the time domain. Then, it will analyze the periodic nesting phenomenon of SST in the tropical western Indian Ocean using EEMD and Fourier transform in the frequency domain. Next, it will use wavelet analysis to study the periods of precipitation, sea surface heat flux, total cloud cover, and longwave radiation, comparing these with SST periods. Finally, it will combine cross-wavelet transform to analyze the relationship between these four factors and SST in the tropical western Indian Ocean at different frequency bands and time ranges. Moreover, this invention employs multiple research methods and combines them with a long and recent time range, enabling a comprehensive analysis of the characteristics of SST changes in the tropical western Indian Ocean over the past 50 years. Summary of the Invention

[0003] The problem this invention aims to solve is to develop a more targeted method for analyzing sea surface temperature (SST) influencing factors based on the climate conditions of the tropical western Indian Ocean. This method analyzes the interannual variation of the SST sequence over the past 47 years using moving averages, moving t-tests, and the MK mutation test. Furthermore, it decomposes the SST sequence into multiple fixed-frequency components using the EEMD method and determines whether the SST sequence is stable by examining the monotonicity of the long-term trend variation term.

[0004] This invention is implemented as follows:

[0005] This invention provides a method for analyzing the influencing factors of sea surface temperature in the tropical western Indian Ocean based on cross wavelets. The implementation process includes the following steps:

[0006] Step 1: First, the temporal variation of the sea surface temperature series in the tropical western Indian Ocean over the past 47 years was analyzed using moving average, moving t-test, and MK mutation test.

[0007] Step 2: Then, the sea surface temperature series is decomposed into multiple fixed-frequency components using EEMD, and the monotonicity of the long-term trend change term is used to determine whether the sea surface temperature series is stationary.

[0008] Step 3: Then, perform a Fast Fourier Transform on the sea surface temperature sequences after high-pass, low-pass, and filtering conditions to obtain the meaningful periods under each condition. At the same time, analyze the hidden multi-period nesting phenomenon of the sea surface temperature sequences from the frequency domain.

[0009] Step 4: In terms of the analysis of influencing factors, four parameters were first selected: precipitation, sea surface heat flux, total cloud cover, and longwave radiation in the tropical western Indian Ocean region. Then, the periodic nesting phenomenon among these four parameters was analyzed using Morlet wavelets, and the order of their main cycles was determined by the magnitude of the wavelet variance. Finally, the correlation between precipitation and sea surface temperature, sea surface heat flux and sea surface temperature, total cloud cover and sea surface temperature, and longwave radiation and sea surface temperature were analyzed using cross wavelets to obtain the influence of these four parameters on sea surface temperature in the tropical western Indian Ocean. The Morlet mother wavelet used in this invention is expressed as Equation (1).

[0010]

[0011] In equation (1), ω is a dimensionless frequency, which is set to a constant of 6 in this paper. When the wavelet scale and the Fourier period are approximately equal, the expression for the continuous wavelet transformation is equation (2).

[0012]

[0013] In equation (2), a is the period length, b is the translation time, a and b ∈ R, and a ≠ 0; the wavelet variance expression is given by equation (3).

[0014]

[0015] Finally, after removing the four factors, one ocean index was selected for influencing factor analysis. From the perspectives of trend changes and correlations, it was concluded that the western Pacific warm pool and the sea surface temperature of the tropical western Indian Ocean are closely related. Analysis from a time-frequency domain perspective provides a more comprehensive understanding of the sea surface temperature variation process in the tropical western Indian Ocean. Attached Figure Description

[0016] Figure 1 This is the overall analysis flowchart of the present invention.

[0017] Figure 2 Figure (a) shows the temporal variation of sea surface temperature in the tropical western Indian Ocean from 1974 to 2020. Figure (b) shows the results of the 8-year moving average t-test analysis of the sea surface temperature series. Figure (c) shows the 9-year moving average test. Figure (d) shows the MK abrupt change test of the sea surface temperature series over the past 47 years.

[0018] Figure 3 This is an EMD decomposition diagram.

[0019] Figure 4 This is a cross-wavelet analysis diagram of precipitation and sea surface temperature in the tropical western Indian Ocean. Detailed Implementation

[0020] The method of the invention will now be described in more detail and clearly with reference to the accompanying drawings.

[0021] like Figure 1 , 2 As shown in Figures 3 and 4, the interannual variation of the sea surface temperature (SST) series in the tropical western Indian Ocean over the past 47 years was analyzed using moving average, moving t-test, and MK mutation test. At the same time, the SST series was decomposed into multiple fixed-frequency components using EEMD, and the monotonicity of the long-term trend variation term was used to determine whether the SST series was stationary.

[0022] First, the interannual variation of the sea surface temperature series in the tropical western Indian Ocean over the past 47 years was analyzed using moving average, moving t-test, and MK mutation test.

[0023] Then, the sea surface temperature series is decomposed into multiple fixed-frequency components using EEMD, and the monotonicity of the long-term trend change term is used to determine whether the sea surface temperature series is stationary.

[0024] Then, a fast Fourier transform was performed on the sea surface temperature sequences after high-pass, low-pass, and filtering conditions to obtain the meaningful periods under each condition. At the same time, the hidden multi-period nesting phenomenon of the sea surface temperature sequences was analyzed from the frequency domain.

[0025] In analyzing the influencing factors, four parameters were first selected: precipitation, sea surface heat flux, total cloud cover, and longwave radiation in the tropical western Indian Ocean region. Then, the periodic nesting phenomenon among these four parameters was analyzed using Morlet wavelets, and the order of their main cycles was determined by the magnitude of the wavelet variance. Finally, the correlation relationships between precipitation and sea surface temperature (SST), sea surface heat flux and SST, total cloud cover and SST, and longwave radiation and SST were analyzed using cross wavelets, revealing the influence of these four parameters on SST in the tropical western Indian Ocean. This invention uses the Morlet mother wavelet, expressed as equation (1).

[0026]

[0027] In equation (1), ω is a dimensionless frequency, which is set to a constant of 6 in this paper. When the wavelet scale and the Fourier period are approximately equal, the expression for the continuous wavelet transformation is equation (2).

[0028]

[0029] In equation (2), a is the period length, b is the translation time, a and b ∈ R, and a ≠ 0; the wavelet variance expression is given by equation (3).

[0030]

[0031] Finally, after removing the four factors, one ocean index was selected for influencing factor analysis. From the perspectives of trend changes and correlations, it was concluded that the western Pacific warm pool and the sea surface temperature of the tropical western Indian Ocean are closely related. Analysis from a time-frequency domain perspective provides a more comprehensive understanding of the sea surface temperature variation process in the tropical western Indian Ocean.

Claims

1. A method for analyzing the influencing factors of sea surface temperature in the tropical western Indian Ocean based on cross wavelets, characterized in that, Includes the following steps: (1) The temporal variation of the sea surface temperature sequence in the tropical western Indian Ocean over the past 47 years was analyzed by moving average, moving t-test and MK mutation test. (2) Then, the sea surface temperature series is decomposed into multiple fixed frequency components by EEMD, and the monotonicity of the long-term trend change term is used to determine whether the sea surface temperature series is stable. (3) Then, perform fast Fourier transform on the sea surface temperature sequence after high-pass, low-pass and filtering conditions to obtain the meaningful period under each condition, and analyze the hidden multi-period nesting phenomenon of the sea surface temperature sequence from the frequency domain. (4) In terms of the analysis of influencing factors, four parameters were first selected: precipitation, sea surface heat flux, total cloud cover and longwave radiation in the tropical western Indian Ocean region. Then, the periodic nesting phenomenon among these four parameters was analyzed by Morlet wavelet analysis, and the main period order of their changes was determined by the size of the wavelet variance. Finally, the correlation between precipitation and sea surface temperature, sea surface heat flux and sea surface temperature, total cloud cover and sea surface temperature, and longwave radiation and sea surface temperature were analyzed by cross wavelet analysis to obtain the influence of these four parameters on sea surface temperature in the tropical western Indian Ocean.

2. The method for analyzing the influencing factors of sea surface temperature in the tropical western Indian Ocean based on cross wavelet as described in claim 1, characterized in that: First, the interannual variation of the sea surface temperature series in the tropical western Indian Ocean over the past 47 years was analyzed using moving average, moving t-test, and MK mutation test.

Images