CN111695250B - A Method for Extracting Internal Tidal Wave Feature - Google Patents

Abstract

The invention discloses a method for extracting internal tidal wave features, which specifically includes preprocessing the ocean reanalysis data, and verifying the reliability of internal tidal wave information extracted from the reanalysis data in the research area by comparing the reanalysis data with the actual measurement data. According to the change of isotherm with time, the research time period is divided, and the characteristic information of internal tidal wave is extracted by using the fluctuation range data of isotherm. The reanalysis data adopted in the present invention has the advantages of wide geographical coverage, high spatial and temporal resolution, etc., through comparison with the actual measurement data, the reanalysis data can better reflect the internal tidal wave situation in the research area, and the research time According to the fluctuation of the underwater isotherm, the section is divided more carefully, and the characteristic information of the internal tidal wave phenomenon in the divided time period is extracted, which more accurately reflects the time variation characteristics of the internal tidal wave.

Description

A Method for Extracting Internal Tidal Wave Feature

technical field

The invention relates to an internal tidal wave feature extraction method, in particular to an internal tidal wave feature extraction method based on ocean reanalysis data, and belongs to the technical field of marine environment statistical research.

Background technique

The internal tidal wave is an important fluctuation existing in the interior of the ocean and plays an important dynamic role in the ocean. The existence of the internal tidal wave will affect various aspects such as the marine ecological environment, ocean detection, and ocean operations. Internal tidal waves will cause fluctuations in the isothermal surface and isodensity surface in the seawater, thereby affecting the acoustic waves in the ocean, changing the amplitude, direction, and propagation speed of the acoustic signal, which will seriously affect the sonar's detection of underwater objects and the detection of underwater objects. Download newsletter. Large-amplitude internal solitary waves closely related to internal tidal waves will generate strong reciprocating shear flow, which will affect the implementation of some ocean engineering. It is of great significance to study the characteristics of internal tidal waves in the ocean.

Internal tidal wave is an internal wave with astronomical tidal cycle, which is the response of a swirling, density-stratified incompressible fluid in the ocean to small disturbances, and is mainly generated by the interaction of astronomical currents and topography. Domestic and foreign studies on internal tidal waves mainly use satellite observation data, mooring observation data, SAR satellite image data, etc. There are many methods to extract the characteristic information of internal tidal waves. The extraction results of different methods reflect the characteristics of internal tidal waves from different angles. The commonly used methods mainly include: using high-pass filtering, polynomial fitting and other filtering methods to separate barotropic flow and baroclinic currents to extract the diurnal and semi-diurnal cycle characteristics of internal tidal waves; use SAR images to detect and orient internal tidal waves to obtain the spatial distribution characteristics of internal tidal waves; use harmonic analysis to analyze seawater temperature or salinity data Through analysis, the harmonic constants of different parts of the internal tidal wave can be obtained, and thus the proportions of different parts of the tidal wave in the overall internal tidal wave can be obtained; the numerical model is used to simulate the internal tidal wave, and the generation and propagation of the internal tidal wave can be obtained mechanism etc. Due to the difficulty in obtaining ocean observation data, the data required for the study of internal tidal waves are relatively scarce. When using the harmonic analysis method to study internal tidal waves, due to frequency confusion, it is necessary to use long-term, high-resolution data to effectively separate Wavelengths of different frequencies, which further reduces the number of observations that can be used.

Contents of the invention

Aiming at the above-mentioned prior art, the technical problem to be solved by the present invention is to provide an internal ocean reanalysis data-based system that overcomes the shortcomings of small coverage of mooring observation data, low time resolution of satellite observation data, and inability to observe underwater. Tidal wave feature extraction method.

In order to solve the above technical problems, a kind of internal tidal wave feature extraction method of the present invention comprises the following steps:

Step 1: Preprocess the ocean reanalysis data: extract the data values of the research sites at different times from each ocean reanalysis data file to form a time series of reanalysis data;

Step 2: Verify the reliability of the reanalysis data to extract the internal tidal wave characteristic information of the research site;

Step 3: Finely divide the research time period;

Step 4: Calculating the fluctuation amplitude value of the isotherm;

Step 5: Use the isotherm fluctuation value obtained in Step 4 to extract the characteristic information of internal tidal waves.

As a preferred solution of the present invention, the reliability of the internal tidal wave feature information of the research site in the verification and reanalysis data extraction described in step 2 specifically includes:

Step 2.1: Calculate the root mean square error between reanalysis data and mooring observation data

Step 2.2: Comparison of reanalysis data and harmonic analysis results of mooring observation data

Step 2.3: When the root mean square error and the harmonic constant difference meet the reliability judgment conditions, it is judged as reliable.

Beneficial effects of the present invention:

Compared with the traditional extraction of internal tidal wave features, the present invention is characterized by the use of marine reanalysis data for the study of internal tidal waves, which has high temporal and spatial resolution, wide geographic coverage, and sufficient underwater data , which is beneficial to study the temporal variation characteristics of underwater internal tidal waves. Firstly, the feasibility of using reanalysis data in the study area is verified, and then the study time period is divided according to the change of isotherms over time, and the influence of the seasons in the ocean on internal tidal wave phenomena is taken into account, which improves the extraction results accuracy. The harmonic analysis is carried out through the isotherm range data of the reanalysis data, which can more intuitively reflect the characteristics of the main components of the internal tidal wave.

The marine reanalysis data adopted in this patent solves the above problems well. Reanalysis data are data that combine data from multiple different sources. These data can be satellite observation data, mooring observation data, numerical model data, etc. Through the analysis and processing of these data, they are fused into a new complete data set, which is reanalysis data. The ocean reanalysis data is generated through the model, which is a new type of data that integrates data from different sources including historical data. It overcomes the small coverage of mooring observation data, low time resolution of satellite observation data and the inability to observe water In addition, ocean reanalysis data include various ocean elements, such as seawater temperature, seawater salinity, and ocean current velocity. In addition, in different time periods, the position of the isotherm in the ocean has obvious changes. If we directly use the seawater temperature data of a long time series for harmonic analysis to explore internal tidal waves, the result is that the seasonal influence is ignored. Accurate results. In this patent, before the harmonic analysis, the time period is finely divided, and the harmonic analysis is carried out in each sub-time period, and the isotherm fluctuation range data is used for the harmonic analysis, and the obtained results can reflect the internal tidal wave more accurately. The temporal variation characteristics more intuitively reflect the contribution of different tidal waves in the internal tidal wave.

(1) It proves that the reanalysis data can be used to extract the characteristic information of internal tidal waves in areas where internal tidal waves are relatively active, which provides new data support for the study of internal tidal waves.

(2) According to the distribution of underwater isotherms at different times, the time period of the harmonic analysis is finely divided, and the harmonic analysis is carried out in each sub-time period, so that the influence of the ocean season on the internal tidal wave is considered, and the results of the harmonic analysis can better reflect Out of the real situation of tidal waves.

(3) The position of the isotherm is compared with the baseline to obtain the fluctuation range of the isotherm, and the fluctuation range of the isotherm is used as the input data of the harmonic analysis. The result of the harmonic analysis more intuitively reflects the characteristics of the main components of the internal tidal wave.

Description of drawings

Figure 1 is a flow chart of the extraction method of internal tidal wave features based on ocean reanalysis data

detailed description

The invention proposes a method for extracting characteristic information of internal tidal waves based on ocean reanalysis data, so as to realize more accurate and intuitive feature information extraction of internal tidal waves. The specific implementation of this method includes preprocessing the ocean reanalysis data, comparing the processed reanalysis data with the mooring observation data, verifying the reliability of the reanalysis data at the research site to study internal tidal waves, and finely dividing the research time period , in each sub-period, the isotherm fluctuation amplitude data is used as input data for harmonic analysis. The execution flow of the internal tidal wave feature extraction method based on reanalysis data according to the present invention is shown in FIG. 1 . The present invention will be described in further detail below.

The present invention proposes a method for extracting internal tidal wave features based on ocean reanalysis data, which specifically includes the following steps:

Step 1: Preprocessing the marine reanalysis data.

The reanalysis data files used in this patent are named after time, and the longitude range of each file data is from 180° west longitude to 179.9° east longitude, and the latitude range is from 89.95° south latitude to 89.95° north latitude. Extract the data values of the research site at different times from each data file. The main research site of this patent is located in the South China Sea, with latitude and longitude ranging from 99°10′E to 122°10′E, 3°S to 23°27′N between. Due to the difficulty in obtaining observational data for comparison between the South China Sea and the reanalysis data, the location of the data for the reliability verification experiment of the reanalysis data was selected at the equator (0°N, 165°E).

Step 2: Verify the reliability of the reanalysis data to extract the internal tidal wave characteristic information of the research site. Reliability verification specifically includes the following steps:

Step 2.1 Calculation of root mean square error between reanalysis data and mooring observation data.

Select the reanalysis data and mooring observation data at the same latitude and longitude (0°N, 165°E) and at the same time (0:00 on July 31, 1992), select the underwater temperature variable, and use the two data for each underwater layer The difference between the two data of each underwater layer is obtained by making the temperature difference, and the root mean square error value of the two data of the entire vertical section is calculated by using the difference of each water layer. The root mean square error value calculated in this patent accounts for about 5% of the average seawater temperature value of the vertical water layer.

Step 2.2 The reanalysis data is compared with the harmonic analysis results of the mooring observation data.

Select the same longitude and latitude (0°N, 165°E), the same underwater depth (underwater 150 meters), and the same time period (January to March 1992) reanalyzed seawater temperature data and mooring observation seawater temperature data To conduct a harmonic analysis, the specific steps are as follows:

(1) Determine the time resolution and time length of the input SST data, which can be selected according to the principles of frequency confusion and Rayleigh analysis.

(2) Determine whether the year Y of the sea temperature data is a leap year, and calculate the number of days ID from January 1 of this year to the current date.

(3) Obtain the initial phase angle of each tide, and calculate the values of astronomical variables T, s, h, p, p1 through the year Y of the current data and the number of days ID from this date to January 1st. The calculation formula is as follows:

Among them, y represents the year of sampling data.

(4) Calculate the initial phase angle V of each tide through the values of variables T, s, h, p, p1, and the calculation formula is as follows:

Among them, n=0, 1, 2; n1, n2, n3, n4 and n5 take 0 or positive and negative integers, and have corresponding fixed values for different frequencies of tides.

(5) Take the median value D of the number of days in the time series, and calculate the intersection factor f and the intersection correction angle u through Doodson parameters combined with Y and D.

(6) The tide level equation can be expressed as:

Among them, A ₀ represents the average water surface height of the tide, m represents the number of tidal equinoxes, and σ _i represents the angular velocity of tidal equinoxes.

in,

则

令make

but

make

The coefficient AA of the linear equation system obtained by taking partial derivatives of A _i and B _i can be calculated through the values of f, u, TT, W, and V, and the value B on the right side of the linear equation system equal sign. where W represents the frequency of each equinox, and TT represents the number of hours from 0 to the time of the harmonic analysis.

(7) Solving the linear equations obtained by E in step (6) with respect to partial derivatives of A _i and B _i by Gaussian elimination method, to obtain amplitude harmonic constants and delay angle harmonic constants.

The difference between the harmonic constants of the main tides obtained by this patent through the two kinds of data is less than 10%, and combined with the calculation result of the root mean square error in step 2.1 accounting for the proportion of the average seawater temperature value of the vertical water layer is less than 10%, which proves that the research site It is reliable to use reanalysis data to extract internal tidal waves.

Step 3: Carry out fine division of the research time period, including the following steps:

Step 3.1 Draw the isotherm distribution curve in the overall time range.

The research site of this step is located in the northern part of the South China Sea (20.35°N, 116.8°E). The time resolution of the reanalysis data used in this patent is 3 hours. In order to avoid the drawn isotherm fluctuations being too dense, some data were removed value, the time resolution is processed into 18 hours to draw the isotherm distribution curve.

Step 3.2 divides the overall time period into multiple sub-time periods.

According to the isotherm distribution curve drawn in step 3.1, it can be concluded that in different time ranges, the position of the isotherm at the same temperature has obvious changes. The 20°C isotherm in this patent is from January 2 to February 1992 The average position between the 24th is about 120 meters deep, and the average position between February 25th and March 31st is about 160 meters deep. According to the distribution of the 20°C isotherm position, the overall research time period was divided into multiple sub-time periods.

Step 4: Calculate the fluctuation amplitude value of the isotherm.

This patent extracts the distribution curve of the 20°C isotherm from the isotherms of each temperature in each sub-time period, averages the values of the isotherm at different times in the sub-time period, and obtains the depth average value of the 20°C isotherm. In each sub-period, the difference between the actual position of the isotherm and the average depth is made to obtain the fluctuation amplitude value of the isotherm.

Step 5: Use the isotherm fluctuation value obtained in Step 4 to extract the characteristic information of internal tidal waves, which specifically includes the following steps:

Step 5.1 takes the isotherm fluctuation amplitude value as input data for harmonic analysis.

The isotherm fluctuation amplitude values obtained in step 4 constitute the time series in each sub-time period, and these time series are used as input data for harmonic analysis to obtain the 20°C isotherm caused by the main tide in each sub-time period The amplitude harmonic constant for the amplitude of the fluctuations.

Step 5.2 conducts a comparative analysis of the harmonic constants of the main tides in each time period.

Through the results of the harmonic analysis, it can be concluded that the tidal equinoxes in the research site of this patent are mainly diurnal tidal equinoxes, and the diurnal tidal equinoxes are mainly K1, O1, P1, and Q1 tidal equinoxes. In the four time periods divided by this patent, the proportion of the K1 tidal tide is the largest, except that the amplitude harmonic constant of the Q1 tidal tide is basically unchanged, and the amplitude harmonic constants of the other three tidal tidal periods have obvious alternating magnitudes transformation phenomenon. The characteristic information of the internal tidal wave can be obtained through the variation of the main tidal harmonic constants of the internal tidal wave in the sub-period.

Specific embodiments of the present invention also include:

The invention relates to the technical field of statistical research on the marine environment, and in particular proposes a method for extracting internal tidal wave features based on marine reanalysis data. Specifically, it includes preprocessing the ocean reanalysis data, verifying the reliability of tidal wave information extracted from the reanalysis data in the study area by comparing the reanalysis data with the measured data, and analyzing the research time period according to the change of the isotherm over time. Carry out division, and use the isotherm fluctuation amplitude data to extract the characteristic information of internal tidal waves. In terms of information extraction and analysis of internal tidal waves, the data used by predecessors usually include: satellite observation data, mooring observation data and SAR image data. The present invention uses marine reanalysis data to extract characteristic information of internal tidal wave phenomena. The reanalysis data has the advantages of wide geographical coverage, high spatial and temporal resolution, etc. Through comparison and verification with actual measurement data, the reanalysis data can be better It reflects the internal tidal wave situation in the study area. The present invention divides the research time period more carefully according to the fluctuation of the underwater isotherm, extracts the characteristic information of the internal tidal wave phenomenon in the divided time period, and more accurately reflects the time of the internal tidal wave change characteristics.

A method for extracting internal tidal wave features based on ocean reanalysis data, including the following steps:

Step 1: Preprocessing the marine reanalysis data.

The reanalysis data files used in this patent are named after time, and the longitude range of each file data is from 180° west longitude to 179.9° east longitude, and the latitude range is from 89.95° south latitude to 89.95° north latitude. The data values of the research sites at different times are extracted from each data file to form a reanalysis data time series.

Step 2: Verify the reliability of the reanalysis data to extract the internal tidal wave characteristic information of the research site.

Using the mooring observation data in the same geographical location, select the same research time and research time period, calculate the root mean square error of the measured data and reanalysis data, and reconcile the time series of the two data in the research time period Analysis and comparison of harmonic analysis constants prove that it is reliable to use reanalysis data to extract internal tidal wave information at the research site.

Step 3: Finely divide the research time period.

Analyze the fluctuation of the underwater isotherm in the overall time period, and divide the overall time period into different sub-time periods according to the distribution depth of the same temperature isotherm with time. period.

Step 4: Calculate the fluctuation amplitude value of the isotherm.

The isotherm distribution of a certain temperature is extracted from the isotherm distribution of various underwater temperatures. This patent extracts the 20°C isotherm, and calculates the average position of the 20°C isotherm distribution underwater in different sub-time periods. In each sub-time period, the difference between the actual position of the isotherm and the average position is made to obtain the fluctuation range value of the isotherm.

Step 5: Use the isotherm fluctuation value obtained in Step 4 to extract the characteristic information of internal tidal waves.

Harmonic analysis is carried out on the isotherm fluctuation amplitude value series in each sub-period, and the harmonic constants of the main tidal components of the internal tidal wave are obtained, and the characteristic information of the internal tidal waves is obtained by analyzing the harmonic constants of the main tidal components.

Claims (2)

1. The method for extracting the internal tide features is characterized by comprising the following steps of:

the method comprises the following steps: preprocessing the marine reanalysis data: extracting data values of research sites at different moments from each ocean reanalysis data file to form a reanalysis data time sequence;

step two: verifying and analyzing the reliability of the information of the tide wave characteristic of the data extraction research site;

step three: finely dividing a research time period;

step four: calculating the fluctuation amplitude value of the isotherm, specifically: extracting the isotherm distribution of a certain temperature from the isotherm distribution of various underwater temperatures, calculating the average position of the extracted isotherm of the temperature in different sub-time periods in the underwater distribution, and subtracting the actual position from the average position of the isotherm in each sub-time period to obtain the fluctuation amplitude value of the isotherm;

step five: extracting the characteristic information of the internal tidal wave by using the isotherm fluctuation amplitude value obtained in the step four, which specifically comprises the following steps:

and 5.1, taking the temperature contour fluctuation amplitude value as input data to carry out harmonic analysis:

forming a time sequence in each sub-time period by the isothermal line fluctuation amplitude value obtained in the fourth step, and performing harmonic analysis on the time sequence serving as input data to obtain an amplitude harmonic constant of the extracted temperature isothermal line fluctuation amplitude caused by main tide splitting of the tide in each sub-time period;

step 5.2, the harmonic constants of the main tide divisions in all time periods are compared and analyzed:

and obtaining the characteristic information of the internal tidal wave according to the change condition of the main tide-dividing harmonic constant of the internal tidal wave in the sub-time period.

2. The method for extracting the internal tide feature as claimed in claim 1, wherein: the reliability of the information of the tidal wave characteristic of the verification reanalysis data extraction research site specifically comprises the following steps:

step 2.1: calculating the root mean square error of the reanalysis data and the mooring observation data;

step 2.2: then the harmony analysis result of the analysis data and the mooring observation data is compared;

step 2.3: and when the difference between the root mean square error and the harmonic constant meets the reliability judgment condition, judging the system to be reliable.

Images