Detection Method for the Position, Morphology and Intensity Characteristics of the Jet Stream in the Upper Atmosphere

Abstract

The present invention discloses a method for detecting the position, morphology and intensity characteristics of the upper-level jet stream. The steps are as follows: select a wind speed critical value; set the grid point values with wind speeds less than the critical value within the jet stream area to 0, and the rest to 1, and apply connected region analysis to obtain the high-value region; use the ellipse fitting method to fit the high-value region into an ellipse; calculate the goodness-of-fit index of the fitted ellipse. If the goodness-of-fit of the ellipse for all years exceeds 0.8, then the fitting is passed, otherwise, select the wind speed critical value and fit again; according to the fitted ellipse, obtain seven jet stream indices including the central longitude, central latitude, width, height, area, angle and intensity. These seven indices can characterize the changes in the position, morphology and intensity characteristics of the jet stream. This method is beneficial to more accurately identify the jet stream and has certain practical significance for the study of jet stream changes and their climate effects.

Description

Technical Field

[0001] The present invention relates to a method for detecting upper-air jet characteristics, and particularly to a detection method for obtaining the position, shape and intensity characteristics of upper-air jets based on image recognition and ellipse fitting techniques, belonging to the field of atmospheric science. Background Art

[0002] The upper-air jet is a strong and narrow air current belt around the earth. The maximum wind speed at the center is greater than or equal to 30 m / s in the upper troposphere. The horizontal length of the jet usually reaches tens of thousands of kilometers, the horizontal width is several hundred kilometers, and the vertical thickness is about several kilometers. In the mid-latitudes, the subtropical westerly jet is an important part of the atmospheric circulation system. It plays an important role in the formation and development of mid-latitude cyclones, the formation of mesoscale convective systems, and the stratosphere-troposphere exchange. Various changes in the jet are also potential indicators of climate change.

[0003] The identification of jets is the basis for research on their variations and climate effects. In previous studies, various methods have been used to identify jets. However, the premise of these methods is to first determine the jet active area, which may have certain limitations because this fixed area may not be applicable to all years. For example, in some years, if the jet is strong enough, it will cover a larger area, and the previously defined fixed area may cut off the entire shape of the jet. That is to say, the upper-air jet may have a unique active area based on wind speed distribution, so the identification of jets has become the focus of related research. In addition, existing technologies for the identification and research of jets basically only focus on the changes in the position and intensity of jets, while ignoring the shape changes of jets. But in fact, some related jet conceptual models have confirmed jets with straight, inclined, large-area, small-area and other shapes, and it has been found that these different-shaped jets will lead to some unique weather and climate phenomena. Therefore, studying the shape characteristics of jets will help to understand the physical mechanism and climate effect of jets. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to accurately identify the position, shape and intensity characteristics of upper-air jets.

[0005] To solve the above technical problems, the method for detecting the position, shape and intensity characteristics of the upper-air jet in the present invention includes the following steps: Select a wind speed value as the wind speed critical value for the jet region; Set the values of the grid points with wind speeds less than the wind speed critical value in the jet region to 0, and the rest to 1, and apply connected region analysis to the grid points, and regard a region where all points exceed the wind speed critical value as a high-value region; Fit all high-value regions into an ellipse by the ellipse fitting method; Obtain the goodness-of-fit index according to the ellipse fitting error and the ellipticity index of the area of the high-value region. If the ellipse fitting goodness-of-fit for all years exceeds 0.8, then perform fitting, otherwise re-select a wind speed critical value for re-fitting; According to the fitted ellipse, obtain seven jet indices including central longitude, central latitude, width, height, area, angle and intensity. The central longitude index and the central latitude index characterize the change of the jet position, the width index, the height index, the area index and the angle index characterize the shape characteristics of the jet, and the intensity index characterizes the change of the jet intensity.

[0006] In the above technical solution, the ellipse fitting method is to obtain the contour point set of the high-value region through detection, and then perform ellipse fitting on the contour point set to obtain the fitted ellipse, and return the center point, major axis, minor axis and rotation angle of the fitted ellipse.

[0007] In the above technical solution, the calculation formula of the goodness-of-fit index is: Among them, A represents the area of the high-value region, N represents the number of contour points of the high-value region, and SE is the ellipse fitting error, and its calculation method is: Among them, d i is the approximate value of the distance from a certain contour point P of the high-value region to the ellipse, and is obtained by calculating the Euclidean distance from P to the intersection point I of the confocal orthogonal hyperbola passing through point P and the ellipse.

[0008] In the above technical solution, the central longitude is the longitude of the center point of the fitted ellipse, the central latitude is the latitude of the center point of the fitted ellipse, the width is the actual great circle distance of the major axis of the fitted ellipse, the height is the actual great circle distance of the minor axis of the fitted ellipse, the area is the actual geographical area of the fitted ellipse, the angle is the included angle between the major axis of the fitted ellipse and the meridian, the intensity is the average wind speed in the fitted ellipse region, and the calculation of the great circle distance adopts: r is the radius of the earth, respectively represent the latitudes of two points on the spherical surface, and Δλ is the longitude difference between the two points.

[0009] In the above technical solution, the contour point set of the high-value region is detected by using the findContours() function of the OpenCV library in Python, and then the contour point set is subjected to ellipse fitting by using the fitEllipse() function.

[0010] In the above technical solution, the jet stream region is the 300 hPa troposphere.

[0011] In the above technical solution, the high-value region contains at least 10 grid points.

[0012] The method for detecting the position, shape and intensity characteristics of the upper-air jet stream in this solution proposes a solution for automatically detecting the upper-air jet stream and determining its position, shape and intensity characteristics, which is conducive to more accurately identifying the jet stream and has certain practical significance for the study of jet stream changes and their climate effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 is a flow chart of the method for detecting the position, shape and intensity characteristics of the upper-air jet stream.

[0014] Figure 2 is a conceptual model of the jet stream after ellipse fitting.

[0015] Figure 3 is a schematic diagram of the ellipse fitting results (a) 1981 (b) 1992, where the black solid-line ellipse represents the ellipse obtained by fitting the high-value region of the wind speed.

[0016] Figure 4 is a histogram of the ellipse fitting goodness-of-fit index obtained by fitting the high-value region of the wind speed from 1979 to 2019;

[0017] Figure 5 is a trend chart of the upper-air jet stream indices from 1979 to 2019, where the broken line is the index change sequence and the solid line is the linear regression result. DETAILED DESCRIPTION OF THE INVENTION

[0018] See Figure 1 , the method for detecting the position, shape and intensity characteristics of the upper-air jet stream of the present invention includes the following steps:

[0019] 1. Select the wind speed critical value. According to the climatological mean of the 300 hPa total wind speed, select a wind speed value as the wind speed critical value for the jet stream region over the years.

[0020] 2. Detect the jet stream active area. Set the values of the grid points with wind speeds less than the wind speed critical value at 300 hPa to 0, and the rest to 1. Apply connected component analysis to these grid points, that is, if all points in a region exceed the critical value, then this region is considered "large enough" and regarded as a high-value region, and the high-value region contains at least 10 grid points.

[0021] 3. Elliptic fitting of jet streams. The high-value region of the jet stream is fitted into an ellipse through the elliptic fitting method, and the geometric features of the output ellipse will be used to characterize the characteristics of the jet stream. The above elliptic fitting method mainly uses the OpenCV library in Python. First, the contour point set of the high-value region is detected using the findContours() function therein, and then the fitEllipse() function is used to perform elliptic fitting on the contour point set, returning the center point, major axis, minor axis, and rotation angle of the fitted ellipse.

[0022] 4. Goodness-of-fit test for elliptic fitting. The goodness-of-fit index for elliptic fitting is obtained based on the ellipticity index of the elliptic fitting error and the area of the high-value region. If the goodness-of-fit for all years exceeds 0.8, the fitting effect is considered good; otherwise, a new wind speed critical value is determined and the fitting is performed again according to the above steps 1 to 3. The calculation formula for the goodness-of-fit index of elliptic fitting is: where A represents the area of the high-value region, N is the number of contour points of the high-value region, and SE is the elliptic fitting error, and its calculation method is: where d i is the approximate distance from a certain contour point P in the high-value region to the fitted ellipse, which is obtained by calculating the Euclidean distance from the contour point P to the intersection point I of the confocal orthogonal hyperbola passing through the contour point P and the fitted ellipse.

[0023] 5. Determination of jet stream index. Based on the fitted ellipse, 7 jet stream indices are obtained, namely: central longitude, central latitude, width, height, area, angle, and intensity index. Among them, the central longitude and central latitude indices can characterize the position change of the jet stream, the width, height, area, and angle indices can characterize the morphological characteristics of the jet stream, and the intensity index can characterize the intensity change of the jet stream. The elliptic fitting of the jet stream is as Figure 2 shown, where the tilted ellipse represents the fitted ellipse. The 7 obtained jet stream indices are: the central longitude is the longitude of the center point of the fitted ellipse; the central latitude is the latitude of the center point of the fitted ellipse; the width is the actual great circle distance of the major axis AB of the fitted ellipse; the height is the actual great circle distance of the minor axis CD of the fitted ellipse; the area is the actual geographical area of the fitted ellipse; the angle is the angle between the major axis AB of the fitted ellipse and the meridian, such as Figure 2 the angle θ in; the intensity is the average wind speed within the fitted ellipse region. Among them, the units of the central longitude, central latitude, and angle are °, the units of the width and height are km, the unit of the area is km 2 , and the unit of the intensity is m / s. The calculation of the great circle distance adopts: r is the radius of the earth, respectively represent the latitudes of two points on the spherical surface, and Δλ is the longitude difference between the two points. The actual geographical area of the fitted ellipse and the average wind speed within the fitted ellipse region can be directly calculated after the fitted ellipse is determined.

[0024] Taking the subtropical upper-level jet in the Northern Hemisphere winter from 1979 to 2019 as an example, using the winter mean value of the full wind speed at 300 hPa in the ERA5 reanalysis data, the position, shape and intensity characteristics of the jet are detected and its changes are analyzed.

[0025] Based on the 1979 - 2019 climatology of the full wind speed at 300 hPa, 52 m / s is selected as the wind speed critical value. The values of the grid points with wind speed less than 52 m / s at 300 hPa are set to 0, and the rest are set to 1. The connected region analysis is applied to obtain the high wind speed regions, and the regions with less than 10 grid points are excluded. The outer contour of the high value region is detected and the contour is fitted to an ellipse using the ellipse fitting method. Figure 3 Examples of the high value regions and the fitted ellipses for two years are given.

[0026] Based on the contour points of the high value region and the fitted ellipse, the goodness of fit of the ellipse is calculated. As Figure 4 shown, it can be found that the goodness of fit for all years from 1979 to 2019 exceeds 0.8, indicating a good fitting effect, and this fitting result can be used to determine the jet index.

[0027] According to the characteristics of the ellipse returned by the fitting, 7 jet index values are calculated for each year. By calculating the correlation coefficients between the jet indices obtained based on different datasets and different wind speed critical values, it is found that they are basically all over 0.8, with a good correlation, which increases the credibility of selecting the ERA5 dataset and the 52 m / s wind speed critical value.

[0028] Based on the obtained 7 jet indices, the changes in the position, shape and intensity of the jet are analyzed. As Figure 5 shown, among them, the change characteristics of the position and intensity indices are consistent with previous studies, which verifies the feasibility of this method. At the same time, our scheme makes up for the identification of the jet shape characteristics.

Claims

1. A method for detecting the position, morphology and intensity characteristics of the upper-level jet stream, characterized in that: Select a wind speed value as the wind speed critical value for the jet stream region; set the values of the grid points with wind speeds less than the wind speed critical value in the jet stream region to 0, and the rest to 1. Apply connected component analysis to the grid points, and regard a region where all points exceed the wind speed critical value as a high-value region; fit all high-value regions into an ellipse by the ellipse fitting method; obtain the goodness-of-fit index according to the ellipse fitting error and the ellipticity index of the area of the high-value region. If the ellipse fitting goodness-of-fit for all years exceeds 0.8, then pass the fitting, otherwise reselect a wind speed critical value for re-fitting; according to the ellipse obtained by fitting, obtain seven jet stream indices: central longitude, central latitude, width, height, area, angle, and intensity. The central longitude index and the central latitude index characterize the change in the position of the jet stream, the width index, the height index, the area index, and the angle index characterize the morphological characteristics of the jet stream, while the intensity index characterizes the change in the intensity of the jet stream.

2. The method for detecting the position, shape and intensity characteristics of the upper-air jet stream according to claim 1, wherein: The ellipse fitting method is to detect and obtain the contour point set of the high-value region, and then perform ellipse fitting on the contour point set to return the center point, major axis, minor axis, and rotation angle of the fitted ellipse.

3. The method for detecting the position, morphology and intensity characteristics of the upper-level jet stream according to claim 1, wherein: The calculation formula of the goodness-of-fit index is as follows: Wherein, A represents the area of the high-value region, N represents the number of contour points of the high-value region, and SE is the ellipse fitting error, and its calculation method is: Wherein, d i is the approximate distance from a certain contour point P of the high-value region to the ellipse, and is obtained by calculating the Euclidean distance from the contour point P to the intersection point I of the confocal orthogonal hyperbola passing through the contour P and the ellipse.

4. The method for detecting the position, shape and intensity characteristics of the upper-air jet stream according to claim 2, wherein: The central longitude is the longitude of the center point of the fitted ellipse, the central latitude is the latitude of the center point of the fitted ellipse, the width is the great circle distance of the major axis of the fitted ellipse, the height is the great circle distance of the minor axis of the fitted ellipse, the area is the actual geographical area of the fitted ellipse, the angle is the angle between the major axis of the fitted ellipse and the meridian, the intensity is the average wind speed within the fitted ellipse area, and the calculation of the great circle distance adopts: r is the radius of the earth, respectively represent the latitudes of two points on the spherical surface, and Δλ is the longitude difference between the two points.

5. The method for detecting the position, shape and intensity characteristics of the upper-level jet according to claim 2, wherein: Use the findContours() function in the OpenCV library in Python to detect and obtain the contour point set of the high-value region, and then use the fitEllipse() function to perform ellipse fitting on the contour point set.

6. The method for detecting the position, morphology and intensity characteristics of the upper-air jet according to claim 1 or 2 or 3 or 4 or 5, characterized in that: The jet stream region is the 300 hPa troposphere.

7. The method for detecting the position, shape and intensity characteristics of the upper-air jet according to claim 1 or 2 or 3 or 4 or 5, characterized in that: The high-value region contains at least 10 grid points.

Images