A method for joint estimation of differential phase and attenuation for dual-polarimetric weather radar

Abstract

The application discloses a differential phase and attenuation joint estimation method for dual-polarized weather radar, improves the precision of phase and attenuation estimation by adaptively adjusting gamma H and gamma DP under high resolution, and reduces the deviation caused by delta HV and noise. The differential phase and attenuation joint estimation method can be applied to dual-polarized weather radar of any wave band, especially high frequency wave bands such as X, Ku and Ka wave bands, and the improvement effect is more remarkable. The application improves the precision of differential phase estimation and attenuation correction, and helps to improve the accuracy of quantitative precipitation estimation. In addition, the application does not introduce new parameters, only adaptively adjusts the relationship formula coefficient between attenuation and differential phase shift rate, and therefore can be applied to radar DSP to realize real-time processing of data.

Description

Technical Field

[0001] This invention relates to the field of weather radar technology, and in particular to a method for joint estimation of differential phase and attenuation for dual-polarization weather radar. Background Technology

[0002] Dual-polarization weather radar is an important tool for quantitative precipitation estimation, and it measures the reflectivity factor Z. H The differential reflectivity Z between orthogonally polarized radar echoes DR and differential phase shift rate K DP These are all important parameters used in rainfall inversion. Therefore, in order to make accurate quantitative precipitation estimates, it is essential to ensure that Z... H Z DR and K DP Accuracy is necessary.

[0003] Differential phase shift rate K DP By propagating the phase Ф through differential propagation DP The distance derivative is obtained, where Ф DP This is the phase difference between the horizontally and vertically polarized echoes caused by forward propagation. However, the measured differential phase Ψ DP Not only includes Ф DP It also includes the backscattering differential phase δ caused by Mie scattering from large droplets or melting ice particles. hv When the frequency is high, δ hv This is particularly evident, and the degree of variation with distance is quite large, especially for K. DP Accurate estimation poses a challenge. Furthermore, in light to moderate rain, Ψ DP There may be noticeable noise, which will also affect K. DP The accuracy of the estimate. Therefore, in order to obtain an accurate K... DP Filter out Ψ DP Noise and δ hv Quantity is necessary.

[0004] Radar signal attenuation affects the measurement of Z. H and Z DR Key factors affecting accuracy. As radar signal beams traverse the atmosphere, they lose energy due to scattering and absorption by precipitation particles. This attenuation primarily depends on the radar pulse wavelength relative to the particle size. At higher frequencies and shorter wavelengths, the attenuation of the radar signal becomes significant. Furthermore, since precipitation particles are not isotropically distributed, the differences in attenuation of orthogonally polarized echoes are also significant. Therefore, to reduce the impact of attenuation on quantitative precipitation estimation, the Z... H and Z DR Accurate attenuation correction is also necessary.

[0005] Because of KDP Unaffected by power attenuation, it is widely used in attenuation correction. However, the accuracy of attenuation correction is affected by K. DP The impact of the accuracy of the estimate, especially at higher frequencies, on δ hv The presence of numerous interfering factors increases the difficulty of accurately estimating K. DP The difficulty. For K DP The estimation, compared to traditional fitting, utilizes the attenuation-corrected Z... H and Z DR Differential phase estimation can significantly improve the resolution and accuracy of the results. Therefore, it is easy to see that, since attenuation is not negligible at higher frequencies, the accuracy of attenuation correction, in turn, affects K. DP Accurate estimation. Furthermore, the attenuation rate A H and differential attenuation rate A DP and differential phase shift rate K DP The empirical coefficient γ in the quasi-linear relationship between them H and γ DP It is also an important factor affecting the accuracy of attenuation correction and differential phase estimation. Because γ H and γ DP It is sensitive to changes in temperature, droplet distribution (DSD), and droplet size, with a wide range of variations, even within the same rainfall area. Therefore, if γ... H and γ DP Setting a fixed prior value will reduce the accuracy of the results, especially when K... DP When γ is large, H and γ DP Inaccuracies can amplify biases. Therefore, adaptive adjustment of γ is necessary at high resolution using the distance library as the basic unit. H and γ DP and reduce δ hv It is essential to mitigate the effects of interference factors to improve the accuracy of attenuation correction and differential phase estimation. Summary of the Invention

[0006] This invention provides a method for joint estimation of differential phase and attenuation for dual-polarization weather radar. The main technical problem it solves is: adaptively and with high resolution adjusting empirical coefficients.

[0007] To address the aforementioned technical problems, this invention provides an adaptive and high-resolution joint estimation method for differential phase and attenuation, comprising:

[0008] Step 1, for radar measurements of Ψ DP The curve is linearly fitted using a sliding window, and the fitted Ψ DP Record

[0009] Step 2, based on the theoretical simulation calculation of γ H and γ DP The range is pre-defined by several different (γ) H ,γ DP Yes, in different (γ) H ,γ DP Under the condition that, based on The first attenuation correction is performed, and the corrected reflectivity factor and differential reflectivity are denoted as follows: and and utilize and as well as Perform the first differential phase estimation, and denote the estimated differential phase shift rate as...

[0010] Step 3, continue at different (γ) H ,γ DP Under the condition that, based on A second attenuation correction is performed using Method I (direct estimation using the empirical relationship between attenuation and phase) and Method II (constraining the attenuation estimate using phase). Furthermore, for each range library, an attenuation rate A is sought that satisfies both methods' estimation. H and differential attenuation rate A DP The smallest difference (γ) H ,γ DP ) is the optimal (γ) H ,γ DP ) and the optimal (γ) H ,γ DP The estimated value of method II under the given conditions is denoted as and

[0011] Step 4, achieving optimality in each distance library (γ) H ,γ DP Under the correct conditions, utilize After performing the third attenuation correction, we obtain and This is the final estimate of the attenuation-corrected reflectivity factor and differential reflectivity, and it is used... and as well as Perform a second differential phase estimation to obtain This is the final estimate of the differential phase shift rate;

[0012] Step 5: Change the radial direction and repeat steps 1 to 4 to complete the attenuation correction and differential phase estimation for all radial directions.

[0013] Step 1 includes:

[0014] Since rainfall remains relatively constant over a certain distance, K within that distance... DP (° / km) can be considered a constant, then Ф DP (°) is a linear function of distance. Because when the frequency is high, δ hv Numerous interfering factors, such as (°), exist, thus affecting the measurement of Ψ. DP The (°) curve is fitted using a sliding window linear fit to eliminate the effects of large phase fluctuations. We will then fit the Ψ curve. DP Record

[0015] Step 2 includes:

[0016] Step 21, First Attenuation Correction:

[0017] γ calculated based on theoretical simulation H and γ DP The range is pre-defined by several different (γ) H ,γ DP Yes. The curve is approximated as Ф DP In each group (γ) H ,γ DP Under the correct conditions, through A H (dB / km) and A DP (dB / km) and K DP The empirical linear relationship between the two is used to perform the first attenuation correction, and the intrinsic reflectivity factor Z estimated by the first attenuation correction is... Hint (dB) and intrinsic differential reflectance Z DRint (dB) are respectively: (r k Indicates the position of the k-th distance library, where k represents any distance library in that radial direction:

[0018]

[0019]

[0020] Here, we will refer to them as follows: and

[0021] Step 22, First differential phase estimation:

[0022] First, it should be noted that all Z in the following calculations Hint and Z DRint Both refer to the first decay correction estimate. and And this step continues in different groups (γ) H ,γ DP They were carried out separately under the conditions of ( ).

[0023] For a distance k, calculate Z within multiple sliding windows containing it. DRint The mean of the standard deviations is used as the standard deviation of the distance from the library k. Within a predetermined interval [L] min L max Select all Δr(r) that satisfy the following equation. m ;r n )(r m ≤r k ≤r n And r m ≠r n ):

[0024]

[0025] For each Δr(r) m ;r n ), its corresponding ΔΨ DP (r m ;r n All of these can be considered as Δδ hv (r m ;r n Negligible ΔФ DP (r m ;r n Furthermore, based on ΔΨ DP (r m ;r n According to K DP With Z Hint and Z DRint The downscaling weight w(r) is calculated using the self-consistent (SC) relationship between the two (Equation (4)). k To estimate each Δr(r) m ;r n K corresponding to ) DP (r k (Equation (5)):

[0026]

[0027]

[0028] in, and r m and r n Z of all distances in the library Hint and Z DRint The mean; Δr0 is the length of a single distance library.

[0029] On the distance library k, for all distinct (γ) H ,γDP ) for each case with different lengths of Δr(r) m ;r n ), calculate all K based on its estimate DP (r k Standard deviation And search for Minimum Δr(r) m ;r n Record the length Δr(r). m ;r n Find all paths corresponding to this length, and calculate all K paths of this length. DP (r k The mean of the values ​​is used as the final estimate of the differential phase shift rate at a distance k from the first differential phase estimation. We denote it as...

[0030] Step 3 includes:

[0031] Step 31, Second Attenuation Correction:

[0032] First, it should be noted that all K in the following calculations... DP Both refer to the values ​​obtained from the first differential phase estimation. And this step continues in different groups (γ) H ,γ DP They were carried out separately under the conditions of ( ).

[0033] Repeat equations (1) to (2) (referred to as Method I) to perform a second attenuation correction, and denote the estimated attenuation rate and the differential attenuation rate at a distance of k as follows: and

[0034] At distance k, when performing a second attenuation correction using another method (referred to as Method II), the minimum Δr(r) is obtained for any length recorded in the previous step. m ;r n The path, using its estimated attenuation rates and differential attenuation rates for the H-polarized and V-polarized channels, are as follows:

[0035]

[0036]

[0037] A DP (r k ) = A H (r k )-A V (r k (8)

[0038] Where, ζ Hatt (r k ) and ζ Vatt (r k All are linear units (mm) 6 m -3 The attenuation reflectivity factor of γ; V It is A V With K DP The coefficients of the linear relationship between them, and γ V =γ H -γ DP .

[0039] For all A estimated by equations (6) to (8) H (r k ) and all A DP (r k The final estimated values ​​of the attenuation rate at distance k and the differential attenuation rate based on method II in the second attenuation correction are obtained by averaging the values ​​of each, and are denoted as follows: and

[0040] Step 32, optimal (γ) H ,γ DP The correct choice:

[0041] On the distance library k, for all preset (γ) H ,γ DP Yes, select the (γ) that minimizes the difference between the results of attenuation correction method I and attenuation correction method II (Equation (9)). H ,γ DP ), as the distance library k corresponding to (γ) H ,γ DP ) pair, also known as optimal (γ) H ,γ DP (Equation (10)):

[0042]

[0043]

[0044] Using equations (9) to (10) for all distance libraries in the radial direction, find the optimal (γ) for each distance library. H ,γ DP Yes. Determine if it is in the optimal (γ) state. H ,γ DP ) Corresponding time and The attenuation correction results at this time are respectively denoted as and

[0045] Step 4 includes:

[0046] Step 41, Third Attenuation Correction:

[0047] Will and Substitute into equations (3) to (5) and search again for the formula. Minimum Δr(r) m ;r n ) and its corresponding path. It is optimal in each distance library (γ). H ,γ DP Under the condition of ), repeat equations (6) to (8) to perform the third attenuation correction. Slightly different is that, due to the optimal (γ) of each distance library... H ,γ DP Therefore, equations (6) and (7) should be slightly modified as follows:

[0048]

[0049]

[0050] Note the K here. DP It still refers to We denote the estimated values ​​of the intrinsic reflectivity factor and the intrinsic differential reflectivity at a distance k as follows: and This is the final estimate of the intrinsic reflectivity factor and intrinsic differential reflectivity at a distance k from the source in this invention.

[0051] Step 42, Second differential phase estimation:

[0052] Will and Substitute into equations (3) to (5) and search for new ones. Minimum Δr(r) m ;r n ), and calculate all K values ​​under that length. DP (r k The mean of () is denoted as K. DP2nd (r k ), which is the final estimated value of the differential phase shift rate at a distance k from the source in this invention.

[0053] The beneficial effects of this invention are:

[0054] Because of K DP Unaffected by power attenuation, it is widely used in attenuation correction. However, the accuracy of attenuation correction is affected by K. DP The impact of the accuracy of the estimate, especially at higher frequencies, on δ hv The presence of numerous interfering factors increases the difficulty of accurately estimating K. DP The difficulty. For KDP The estimation, compared to traditional fitting, utilizes the attenuation-corrected Z... H and Z DR Differential phase estimation can significantly improve the resolution and accuracy of the results. Therefore, it is easy to see that, since attenuation is not negligible at higher frequencies, the accuracy of attenuation correction, in turn, affects K. DP Accurate estimation. Furthermore, the attenuation rate A H and differential attenuation rate A DP and differential phase shift rate K DP The empirical coefficient γ in the quasi-linear relationship between them H and γ DP It is also an important factor affecting the accuracy of attenuation correction and differential phase estimation. Because γ H and γ DP It is sensitive to changes in temperature, droplet distribution (DSD), and droplet size, with a wide range of variations, even within the same rainfall area. Therefore, if γ... H and γ DP Setting a fixed prior value will reduce the accuracy of the results, especially when K... DP When γ is large, H and γ DP Inaccuracies can amplify biases. This paper proposes a joint differential phase and attenuation estimation method for dual-polarization weather radar, which adaptively adjusts γ at high resolution. H and γ DP To improve the accuracy of phase and attenuation estimation and reduce δ hv This invention addresses the biases introduced by noise. By applying this invention, attenuation correction and differential phase estimation are performed on data from typical storm events observed by the Dutch polarimetric X-band weather radar, improving the reliability of the data. Attached Figure Description

[0055] Figure 1 This is a schematic diagram of the differential phase and attenuation joint estimation method for dual-polarization weather radar according to Embodiment 1 of the present invention.

[0056] Figure 2 This is a measured reflectance factor diagram of Embodiment 1 of the present invention;

[0057] Figure 3 This is a measured differential reflectance diagram of Embodiment 1 of the present invention;

[0058] Figure 4 This is a measured differential propagation phase diagram of Embodiment 1 of the present invention;

[0059] Figure 5 This is a reflectance factor diagram corrected after adaptively and at high resolution adjusting the empirical coefficients according to Embodiment 1 of the present invention.

[0060] Figure 6 This is a differential reflectance map corrected after adaptively and with high resolution adjusting the empirical coefficients according to Embodiment 1 of the present invention.

[0061] Figure 7 This is a differential phase shift rate map estimated after adaptively and with high resolution adjusting the empirical coefficients according to Embodiment 1 of the present invention.

[0062] Figure 8 This is the differential propagation phase diagram estimated after adaptively and with high resolution adjusting the empirical coefficients according to Embodiment 1 of the present invention;

[0063] Figure 9 This is a reflectance factor diagram corrected using prior fixed empirical coefficients, as shown in Embodiment 1 of the present invention.

[0064] Figure 10 This is a differential reflectance map corrected using prior fixed empirical coefficients, as shown in Embodiment 1 of the present invention.

[0065] Figure 11 This is a differential phase shift rate diagram estimated using prior fixed empirical coefficients in Embodiment 1 of the present invention;

[0066] Figure 12 This is the differential propagation phase diagram estimated using prior fixed empirical coefficients in Embodiment 1 of the present invention;

[0067] Figure 13 The stability graph of the reflectivity factor within the area enclosed by the black circle over three minutes after adaptively and at high resolution adjusting the empirical coefficients and using a priori fixed empirical coefficients, respectively, is shown in Embodiment 1 of the present invention.

[0068] Figure 14 The stability graph of the differential reflectance within the area enclosed by the black circle over three minutes after adaptively and at high resolution adjusting the empirical coefficients and using the prior fixed empirical coefficients, respectively, is shown in Embodiment 1 of the present invention.

[0069] Figure 15 This is a stability graph of the differential phase shift rate within the region enclosed by the black circle over three minutes after adaptively and at high resolution adjusting the empirical coefficients and using the prior fixed empirical coefficients, respectively, according to Embodiment 1 of the present invention. Detailed Implementation

[0070] To make the objectives and technical solutions of this invention clearer, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0071] Example 1:

[0072] In this embodiment, data from the Polarimetric X-band (9.45 GHz) Weather Radar (IDRA) at the International Centre for Telecommunications and Radar Research (IRCTR) observation station in Cabauw, Netherlands (NL) is used to verify the content of this invention. The radar is 213 m above the ground, with an operating range and range resolution of 15.3 km and 0.03 km, respectively. The scanning height is fixed at 0.50°, the antenna rotates 360° within 1 minute, and the beamwidth is 1.8°. A convective storm event occurred in the Netherlands on September 10, 2011. For the 1946 UTC data measured by the radar, Z... Hatt Z DRatt Ψ DP The linear depolarization ratio L DR A distance library (dB) greater than -18dB was used to remove particles other than rain and / or regions with low signal-to-noise ratio (SNR). Furthermore, to reduce the impact of noise, Z... Hatt Z DRatt Ψ DP Speckle filtering was performed separately. The PPI values ​​are as follows: Figures 2-4 As shown.

[0073] This embodiment provides a method for joint estimation of differential phase and attenuation for dual-polarization weather radar. Please refer to [link to relevant documentation]. Figure 1 Specifically, it includes the following steps:

[0074] Step 1 specifically includes:

[0075] Since rainfall remains relatively constant over a certain distance (usually set at 3km), K within this distance... DP (° / km) can be considered a constant, then Ф DP (°) is a linear function of distance. Because when the frequency is high, δ hv Numerous interfering factors, such as (°), exist, thus affecting the measurement of Ψ. DP The (°) curve is fitted with a linear fit using a 3km sliding window to eliminate the effects of large phase fluctuations. We will then fit the Ψ curve. DP Record

[0076] Step 2 specifically includes:

[0077] Step 21, First Attenuation Correction:

[0078] Based on the theoretical simulation results of X-band radar, for multiple preset sets of different (γ) H ,γ DP Yes, [γ] Hmin ,γ Hmax The preset value is [0.137, 0.337] (dB°).-1 ), and with 0.02 (dB°) -1 ) as step size; [γ DPmin ,γ DPmax The preset value is [0.013, 0.063] (dB°). -1 ), and with 0.01 (dB° -1 ) is used as the step size. There are a total of 11 × 6 = 66 types (γ) H ,γ DP )right.

[0079] Will The curve is approximated as Ф DP In each group (γ) H ,γ DP Under the correct conditions, through A H (dB / km) and A DP (dB / km) and K DP The empirical linear relationship between the two is used to perform the first attenuation correction, and the intrinsic reflectivity factor Z estimated by the first attenuation correction is... Hint (dB) and intrinsic differential reflectance Z DRint (dB) are respectively: (r k Indicates the position of the k-th distance library, where k represents any distance library in that radial direction:

[0080]

[0081]

[0082] Here, we will refer to them as follows: and

[0083] Step 22, First differential phase estimation:

[0084] First, it should be noted that all Z in the following calculations Hint and Z DRint Both refer to the first decay correction estimate. and And this step continues in different groups (γ) H ,γ DP They were carried out separately under the conditions of ( ).

[0085] For a distance k, calculate Z within multiple sliding windows containing it. DRint The mean of the standard deviations is used as the standard deviation of the distance from the library k. Based on the results of theoretical simulation of X-band radar, the predetermined Δr(r) m ;r nThe length interval of Δr(r) is [3km; 5km], with a step size of 0.03km. Within this interval, select all Δr(r) that satisfy the following formula. m ;r n )(r m ≤r k ≤r n And r m ≠r n ):

[0086]

[0087] For each Δr(r) m ;r n ), its corresponding ΔΨ DP (r m ;r n All of these can be considered as Δδ hv (r m ;r n Negligible ΔФ DP (r m ;r n Furthermore, based on ΔΨ DP (r m ;r n According to K DP With Z Hint and Z DRint The downscaling weight w(r) is calculated using the self-consistent (SC) relation (Equation (16)). k To estimate each Δr(r) m ;r n K corresponding to ) DP (r k (Equation (17)):

[0088]

[0089]

[0090] In the X-band, c1 = 1.37 × 10⁻⁶ -3 c2 = 0.68 and c3 = -0.042; and r m and r n Z of all distances in the library Hint and Z DRint The mean value; Δr0 = 0.03km is the length of a single distance reservoir.

[0091] On the distance k, for each of the 66 cases, Δr(r) is a different length. m ;r n ), calculate all K based on its estimateDP (r k Standard deviation And search for Minimum Δr(r) m ;r n Record the length Δr(r). m ;r n Find all paths corresponding to this length, and calculate all K paths of this length. DP (r k The mean of ) is used as the final K distance from the library k. DP Here we will refer to it as

[0092] Step 3 specifically includes:

[0093] Step 31, Second Attenuation Correction:

[0094] First, it should be noted that all K in the following calculations... DP Both refer to the values ​​obtained from the first differential phase estimation. And this step continues in 66 different (γ) H ,γ DP They were carried out separately under the conditions of ( ).

[0095] Repeat equations (13) to (14) (referred to as Method I) to perform a second attenuation correction, and denote the estimated attenuation rate and the differential attenuation rate at a distance of k as follows: and

[0096] At distance k, when performing a second attenuation correction using another method (referred to as Method II), the minimum Δr(r) is obtained for any length recorded in the previous step. m ;r n The path, using its estimated attenuation rates and differential attenuation rates for the H-polarized and V-polarized channels, are as follows:

[0097]

[0098]

[0099] A DP (r k ) = A H (r k )-A V (r k (20)

[0100] Where, ζ Hatt (r k ) and ζ Vatt (r k All are linear units (mm)6 m -3 The attenuation reflectivity factor of γ; V It is A V With K DP The coefficients of the linear relationship between them, and γ V =γ H -γ DP Because the coefficient b H and b V The range of variation for both is very small, therefore both are set to constant values, i.e., b. H ≡0.7644, b V ≡0.7968.

[0101] For all A estimated by equations (18) to (20) H (r k ) and all A DP (r k The final estimated values ​​of the attenuation rate at distance k and the differential attenuation rate based on method II in the second attenuation correction are obtained by averaging the values ​​of each, and are denoted as follows: and

[0102] Step 32, optimal (γ) H ,γ DP The correct choice:

[0103] On the distance library k, for 66 preset (γ) H ,γ DP Yes, select the (γ) that minimizes the difference between the results of attenuation correction method I and attenuation correction method II (Equation (21)). H ,γ DP ), as the distance library k corresponding to (γ) H ,γ DP ) pair, also known as optimal (γ) H ,γ DP For (Equation (22)):

[0104]

[0105]

[0106] Using equations (21) to (22) for all distance libraries in the radial direction, find the optimal (γ) for each distance library. H ,γ DP Yes. Determine if it is in the optimal (γ) state. H ,γ DP ) Corresponding time and The attenuation correction results at this time are respectively denoted as and

[0107] Step 4 specifically includes:

[0108] Step 41, Third Attenuation Correction:

[0109] Will and Substitute into equations (15) to (17) and search again for the formula. Minimum Δr(r) m ;r n ) and its corresponding path. It is optimal in each distance library (γ). H ,γ DP Under the condition of ), repeat equations (18) to (20) to perform the third attenuation correction. Slightly different is that, due to the optimal (γ) of each distance library... H ,γ DP Therefore, equations (18) and (19) should be slightly modified as follows:

[0110]

[0111]

[0112] Note the K here. DP It still refers to We denote the estimated values ​​of the intrinsic reflectivity factor and the intrinsic differential reflectivity at a distance k as follows: and This refers to the final estimated values ​​of the intrinsic reflectivity factor and intrinsic differential reflectivity at a distance k from the source in this invention, with PPI as shown below. Figure 5 and 6 As shown.

[0113] Step 42, Second differential phase estimation:

[0114] Will and Substitute into equations (15) to (17) and search for new ones. Minimum Δr(r) m ;r n ), and calculate all K values ​​under that length. DP (r k The mean of () is denoted as K. DP2nd (r k ), which is the final estimated value of the differential phase shift rate at a distance k from the source in this invention, PPI as follows. Figure 7 As shown. Φ after path integration DP PPI such as Figure 8 As shown.

[0115] Instead of adaptively and with high resolution adjusting γ, a prior fixed value (γ) is used.H and γ DP All take the maximum value, i.e., γ H =0.337, γ DP =0.063), using equations (15) to (17) and (18) to (20) respectively for phase estimation and attenuation correction, the obtained Z Hint Z DRint K DP and Ф DP PPI such as Figures 9-12 As shown.

[0116] To address the adaptive variation of empirical coefficients with different rainfall conditions at high resolution, this paper analyzes the Z-values ​​estimated using two different schemes. Hint Z DRint K DP The stability over time was used to evaluate the performance of the invention. To mitigate the impact of rapid changes in convective storms on the stability assessment, a localized rainfall area in the northeast (circled in black) with moderate intensity during the period of 1945 UTC to 1947 UTC was selected, where rainfall was relatively stable. Furthermore, to reduce the influence of random biases and abrupt changes, the average value was used instead of the Z-value for a single distance pool in this area. Hint Z DRint K DP Comparison, such as Figures 13-15 As shown. It can be seen that, regardless of Z Hint Z DRint K DP This invention provides the best stability. The adaptive and high-resolution joint estimation method for differential phase and attenuation improves the accuracy of the estimates.

[0117] The above description, in conjunction with specific embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A method for joint estimation of differential phase and attenuation for dual-polarization weather radar, characterized in that, include: Step 1, for radar measurements of Ψ DP The curve is linearly fitted using a 3km sliding window, and the fitted Ψ DP Record ; Step 2, based on the theoretical simulation calculation of γ H and γ DP The range is pre-defined by several different (γ) H ,γ DP Yes, in different (γ) H ,γ DP Under the condition that, based on The first attenuation correction is performed, and the corrected reflectivity factor and differential reflectivity are denoted as follows: and and utilize and as well as Perform the first differential phase estimation, and denote the estimated differential phase shift rate as... ; Step 3, continue at different (γ) H ,γ DP Under the condition that, based on The attenuation is corrected using two methods: Method I (direct estimation using the empirical relationship between attenuation and phase) and Method II (constraining the attenuation estimate with phase). Furthermore, for each range library, the attenuation rate estimated by both methods is sought. A H and differential attenuation rate A DP The smallest difference (γ) H ,γ DP ) is the optimal (γ) H ,γ DP ) and the optimal (γ) H ,γ DP The estimated value of the ZPHI method under the given conditions is denoted as... and ; Step 4, achieving optimality in each distance library (γ) H ,γ DP Under the correct conditions, utilize A third attenuation correction was performed, resulting in... and This is the final estimate of the attenuation-corrected reflectivity factor and differential reflectivity, and it is used... and as well as A second differential phase estimation is performed to obtain... This is the final estimate of the differential phase shift rate; Step 5: Change the radial direction and repeat steps 1 to 4 to complete the attenuation correction and differential phase estimation for all radial directions.

2. The differential phase and attenuation joint estimation method for dual-polarization weather radar as described in claim 1, characterized in that, Step 1 includes: Since rainfall remains relatively constant over a certain distance, therefore within that distance... K DP (° / km) can be considered a constant, then Ф DP (°) is a linear function of distance; since δ is higher at higher frequencies... hv Numerous interfering factors, such as (°), exist, thus affecting the measurement of Ψ. DP (°) The curve is fitted using a sliding window linear fit to eliminate the effects of large fluctuations in the phase; we will fit the Ψ... DP Record .

3. The differential phase and attenuation joint estimation method for dual-polarization weather radar as described in claim 2, characterized in that, Step 2 includes: Step 21, First Attenuation Correction: γ calculated based on theoretical simulation H and γ DP The range is pre-defined by several different (γ) H ,γ DP Yes; will The curve is approximated as Ф DP In each group (γ) H ,γ DP Under the correct conditions, through A H (dB / km) and A DP (dB / km) and K DP The empirical linear relationship between them is used to perform the first attenuation correction, and the intrinsic reflectivity factor estimated by the first attenuation correction is... Z Hint (dB) and intrinsic differential reflectance Z DRint (dB) are respectively: ( r k Indicates the position of the k-th distance library, where k represents any distance library in that radial direction: (1)； (2)； Here, we will refer to them as follows: and ; Step 22, First differential phase estimation: First, it should be noted that all calculations below... Z Hint and Z DRint Both refer to the first decay correction estimate. and And this step continues in different groups (γ) H ,γ DP The following were carried out under the respective conditions; For a distance k, calculate the distance within multiple sliding windows it belongs to. Z DRint The mean of the standard deviations is used as the standard deviation of the distance from the library k. ; within a predetermined interval [ L min ; L max Select all Δ values ​​that satisfy the following formula. r ( r m ; r n ()( r m ≤ r k ≤ r n and r m ≠ r n ): (3)； For each Δ r ( r m ; r n ), its corresponding ΔΨ DP ( r m ; r n All of these can be considered as Δδ hv ( r m ; r n Negligible ΔФ DP ( r m ; r n Furthermore, based on ΔΨ DP ( r m ; r n ),according to K DP and Z Hint and Z DRint Downscaling weights calculated using the self-consistent (SC) relationship between the two (Equation (4)). w ( r k To estimate each Δ r ( r m ; r n ) corresponding K DP ( r k (Equation (5)): (4)； (5)； in, and They are respectively r m and r n All distances between the libraries Z Hint and Z DRint The mean; Δ r 0 represents the length of a single distance library; On the distance library k, for all distinct (γ) H ,γ DP ) for different lengths of Δ in each case of the given situation r ( r m ; r n ), calculate all based on its estimate K DP ( r k Standard deviation and search for Minimum Δ r ( r m ; r n Record the length Δ. r ( r m ; r n Find all paths corresponding to that length, and calculate all paths of that length. K DP ( r k The mean of the values ​​is used as the final estimate of the differential phase shift rate at a distance k from the first differential phase estimation. We denote it as... .

4. The differential phase and attenuation joint estimation method for dual-polarization weather radar as described in claim 3, characterized in that, Step 3 includes: Step 31, Second Attenuation Correction: First, it should be noted that all calculations below... K DP Both refer to the values ​​obtained from the first differential phase estimation. And this step continues in different groups (γ) H ,γ DP The following were carried out under the respective conditions; Repeat equations (1) to (2) (referred to as Method I) to perform a second attenuation correction, and denote the estimated attenuation rate and the differential attenuation rate at a distance of k as follows: and ; At distance k, when performing a second attenuation correction using another method (referred to as Method II), the minimum Δ is obtained for any length recorded in the previous step. r ( r m ; r n The path, using its estimated attenuation rates and differential attenuation rates for the H-polarized and V-polarized channels, are as follows: (6)； (7)； (8)； Where, ζ Hatt ( r k ) and ζ Vatt ( r k All are linear units (mm) 6 m -3 The attenuation reflectivity factor of γ; V yes A V and K DP The coefficients of the linear relationship between them, and γ V =γ H -γ DP ; All of the estimates for equations (6) to (8) A H ( r k ) and all A DP ( r k By averaging these values, the final estimates of the attenuation rate at distance k and the differential attenuation rate based on Method II in the second attenuation correction are obtained, denoted as follows: and ; Step 32, optimal (γ) H ,γ DP The correct choice: On the distance library k, for all preset (γ) H ,γ DP Yes, select the (γ) that minimizes the difference between the results of attenuation correction method I and attenuation correction method II (Equation (9)). H ,γ DP ), as the distance library k corresponding to (γ) H ,γ DP ) pair, also known as optimal (γ) H ,γ DP (Equation (10)): (9)； (10)； Using equations (9)~(10) for all distance libraries in the radial direction, find the optimal (γ) for each distance library. H ,γ DP Yes; determine that it is in the optimal (γ) H ,γ DP ) Corresponding time and And the attenuation correction results at this time are respectively denoted as and .

5. The differential phase and attenuation joint estimation method for dual-polarization weather radar as described in claim 4, characterized in that, Step 4 includes: Step 41, Third Attenuation Correction: Will and Substitute equations (3) to (5) and search again for the formula. Minimum Δ r ( r m ; r n ) and its corresponding path; it is optimal in each distance library (γ) H ,γ DP Under the condition of ), repeat equations (6) to (8) to perform the third attenuation correction; the difference is that, due to the optimal (γ) of each distance library H ,γ DP Therefore, equations (6) and (7) should be slightly modified as follows: (11)； (12)； Note that here... K DP It still refers to We denote the estimated values ​​of the intrinsic reflectivity factor and the intrinsic differential reflectivity at a distance k as follows: and This is the final estimated value of the intrinsic reflectivity factor and intrinsic differential reflectivity at a distance k from the source in this invention; Step 42, Second differential phase estimation: Will and Substitute into equations (3) to (5) and search for new ones. Minimum Δ r ( r m ; r n ), and calculate all values ​​under that length. K DP ( r k The mean of () is denoted as K DP2nd ( r k ), which is the final estimated value of the differential phase shift rate at a distance k from the source in this invention.

Images