Regional snow water equivalent inversion method based on vertical combined GNSS receiver

Abstract

The invention relates to the technical field of satellite navigation remote sensing, and relates to a regional snow water equivalent inversion method based on a vertical combined GNSS receiver, which comprises the following steps of: down-sampling the sampling frequency of the receiver and downloading a resource file; executing a GNSS interference snow depth inversion algorithm; a GNSS refraction SWE inversion algorithm is executed in batches; data cleaning and quality inspection; calculating regional snow density for the mask snow depth and the SWE in the same azimuth angle range; and the SWE of the extended range is obtained by using the product of the regional snow density and the snow depth of the all-dimensional angle range. The invention provides a regional scale snow water equivalent inversion method based on a vertical combined GNSS receiver, is used for two vertically combined GNSS receivers, and relates to combined application of two GNSS inversion algorithms, namely an interference snow depth inversion method and a refraction SWE inversion method. According to the method, regional water resources can be better reflected, and the uncertainty of the SWE of the extended range after verification is 60 mm (-10%).

Description

Technical Field

[0001] The invention belongs to the technical field of satellite navigation and remote sensing, and in particular relates to a regional snow water equivalent inversion method based on a vertically combined GNSS receiver. Background Art

[0002] The Global Navigation Satellite System (GNSS) has the characteristics of global availability and wide coverage. A variety of GNSS remote sensing technologies have been developed and widely used in the fields of sea surface wave height, snow depth measurement and soil moisture. Due to its automatic and non-destructive monitoring characteristics, it is suitable for monitoring snow depth and snow water equivalent (SWE) elements in terrain prone to avalanches to avoid human intervention during snowfall. It is safe to use and does not rely on scarce and labor-intensive snow density measurements.

[0003] At present, two algorithms are implemented in the field of GNSS remote sensing to obtain the above two important snow parameters. Specifically, the GNSS interferometric snow depth inversion algorithm is applied to the GNSS receiver placed on a monitoring pole several meters high. This method can obtain the maximum snow depth limit of 50 cm below the antenna height with an uncertainty of about 10%, and the snow depth represents the average value within a few square kilometers; the GNSS refraction SWE inversion algorithm is implemented for the GNSS receiver buried under the snow, and the daily difference in coordinate height is calculated to invert the SWE of the snow layer. However, this value can only reflect the in-situ range of a few square meters. As an important parameter of the water resource model, if the SWE value can represent a larger area, it can be more accurately applied to water resource applications such as calculating snowmelt runoff, reflecting regional water resources and groundwater recharge.

[0004] The direct inversion of SWE parameters in the in-situ range has been verified to be feasible in recent years. Steiner et al. found that the vertical component was significantly correlated with SWE through covariance diagnosis analysis, and successfully estimated SWE using the change of this component. Henkel et al. considered the impact of snow accumulation on the signal to model and estimate SWE, and verified that the SWE estimate was closely matched with the reference SWE value under dry snow conditions. Capelli et al. verified its applicability and accuracy at four other altitude experimental sites. However, there has been no relevant research on expanding the SWE inversion range. Based on this, it is very necessary to develop a method based on vertical combined GNSS equipment to combine the GNSS interferometric snow depth inversion algorithm with the SWE inversion algorithm, use snow density to establish a connection between the two, and expand the SWE representative range through the inversion range with a large snow depth. This is a new attempt and has broad application value. Summary of the invention

[0005] The purpose of the present invention is to provide a regional snow water equivalent inversion method based on a vertically combined GNSS receiver to solve the problem that the inverted SWE represents a too small range.

[0006] The objective of the present invention is achieved through the following technical solutions:

[0007] A regional snow water equivalent inversion method based on a vertically combined GNSS receiver comprises the following steps:

[0008] S1. Base station receiver sampling frequency downsampling and resource file downloading: The sampling interval of the GNSS base station receiver is downsampled to 15s, and the IGS precise ephemeris, IGS precise clock and antenna correction files for each station corresponding to the day are downloaded for the two stations;

[0009] S2. Execute the GNSS interferometric snow depth inversion algorithm. The selected signal frequencies include S1C, S2S and S2Q, excluding S2W. The preset elevation angle range is 5°-30°. The limit range of the signal distance in the LSP diagram is 0 to 8 meters. The daily execution azimuth range is 0-360° without setting a mask. The azimuth range is the direction of the GNSS mobile station. The azimuth angle is centered at the mobile station azimuth and the range is 40°. Two sets of snow depth inversion results are obtained, and one set of snow depth inversion results is obtained for the receiver range under the snow;

[0010] S3. Batch execute the GNSS refraction SWE inversion algorithm to solve the relative positioning baseline between the GNSS base station and the GNSS mobile station. The sampling interval of the GNSS base station observation file ensures that the time matches the GNSS mobile station file; apply the correction file, use the observation equation and the ionosphere-free combination to estimate the floating point solution of the L1+L2+L5 triple frequency through combined filtering, and the starting plane is the location of the mobile antenna under the snow;

[0011] S4, data cleaning and quality check: perform data cleaning and quality check for the calculated SWE and two sets of snow depths;

[0012] S5, calculating the regional snow density for the snow depth and SWE within the same azimuth range, and calculating the regional snow density ρ = SWE / SD using the snow depth and SWE within the same azimuth range;

[0013] S6. Use the product of regional snow density and snow depth in all-round angular range to obtain SWE in the extended range. The density does not change much in a small area, less than 10%, as an intermediate quantity. The snow depth representing the all-round angular range is again converted into SWE to expand the monitoring range of SWE.

[0014] Furthermore, in step S1, the GNSS base station receiver is located on the mast, and the GNSS mobile station receiver is buried underground at a position of 221° of the GNSS base station azimuth. The sampling interval of the GNSS base station receiver is 1 second, and the sampling interval of the GNSS mobile station receiver is 30 seconds.

[0015] Further, step S2 is specifically as follows: excluding the noisiest S2W frequency among the GNSS signal frequencies received by the GNSS base station, setting the altitude angle range to 5°-30°, limiting the maximum distance to 8 meters, and executing the GNSS interferometric snow depth algorithm for not setting an azimuth mask, i.e., 0-360°, and setting an azimuth mask, i.e., 201°-241°.

[0016] Furthermore, in step S2, the inversion of the GNSS interferometric snow depth algorithm uses the phase difference between the reflected signal and the direct signal to reflect the plane height, i.e., the snow depth. The signal-to-noise ratio (SNR) data used is expressed as the signal strength received by the GNSS antenna. In the low altitude angle range, the SNR is specifically the direct signal power P d and the reflected signal power P r Interference between:

[0017]

[0018] Where φ is the phase difference between the two signals. The unit dB of SNR is usually converted to a linear scale. When the direct signal is removed from the fit using a low-order polynomial, the cosine reflected signal dependence on the satellite elevation angle E is obtained:

[0019]

[0020] Among them, S r represents the reflected signal sequence, A is the amplitude, λ is the carrier wavelength, is the phase shift;

[0021] The main frequency f = 2h / λ is obtained using the adaptive Lomb-Scargle periodogram LSP, thereby obtaining the reflection height h; the parameter threshold is set, the peak signal-to-noise ratio is 3, and the primary and secondary signal-to-noise ratio is 1.5. The final SD obtained is the difference between the reflection height and the plane reference surface.

[0022] Further, step S3 is specifically:

[0023] Additional error source δL caused by snow layer m Defined as the excess path length between the receiver under the snowpack and the receiver above the snowpack when positioned relative to each other:

[0024]

[0025] Where λ is the wavelength of the carrier phase; is the phase observation value; satellite pair {k,l}; is the normalized line-of-sight vector between the satellite and receiver antenna phase centers; is the baseline vector between the rod and the ground antenna; all known path delays (from the antenna under the snow to the GPS satellite), including atmospheric delay, integer ambiguity, multipath error and other errors. (·) kl =(·) k -(·) l is the difference between satellites k and l;

[0026] When the baseline solution is performed for the receivers above and below the snowpack without removing the error term, the Up component in the fixed baseline result will change. The SWE is derived from the deviation of the Up component, and the change is shown in the following formula:

[0027] δSWE≈δUp (4)

[0028] RTKLIB b34 is used to solve the relative positioning baseline of the GNSS reference station and the GNSS mobile station; the sampling interval of the reference station observation file is 1 second, using the full elevation angle, as well as the IGS precise ephemeris, IGS precise clock and antenna correction files for each station. The floating-point solution of the L1+L2+L5 triple frequency is estimated by combined filtering using the observation equation and ionosphere free combination. The starting plane is the location of the mobile antenna under the snow. The height difference between the two antennas is 325 cm, which becomes 625 cm after doy39 in 2022.

[0029] Further, step S4 is specifically as follows: counting the valid days of snow depth and SWE obtained under two azimuths, excluding days with different data, and excluding days with excessively large standard deviation of snow depth within the azimuth.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] The present invention provides a regional snow water equivalent inversion method based on a vertically combined GNSS receiver, which is used to improve the problem of too small representative range of the existing SWE inversion algorithm. The method is used for two vertically combined GNSS receivers and is a new strategy involving the combined application of two types of GNSS inversion algorithms - interferometric snow depth inversion method and refraction SWE inversion method. The present invention calculates the density according to the same azimuth range to expand the range of SWE, which can better reflect regional water resources. The verified uncertainty of the SWE in the extended range is 60 mm (~10%). BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0033] Figure 1 It is a schematic diagram of a vertically combined GNSS receiver device according to the present invention;

[0034] Figure 2 It is a flow chart of the regional snow water equivalent inversion method based on the vertical combined GNSS receiver of the present invention;

[0035] Figure 3 This is a comparison chart for evaluating the effects of the methods of the present invention. DETAILED DESCRIPTION

[0036] The present invention will be further described below in conjunction with embodiments:

[0037] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0038] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0039] The invention discloses a regional snow water equivalent inversion method based on a vertically combined GNSS receiver, comprising the following steps:

[0040] S1. Reduce the sampling frequency of the base station receiver and download resource files. The sampling interval of the GNSS base station receiver located on the mast is reduced to 15s. Download the IGS precise ephemeris, IGS precise clock and antenna correction files for each station for the day for the two stations.

[0041] S2. Execute the GNSS interferometric snow depth inversion algorithm, and the selected signal frequencies include S1C, S2S and S2Q, excluding S2W. The preset altitude angle range is 5°-30°. The limit range of the signal distance in the LSP diagram is 0 to 8 meters. The daily execution azimuth range is no mask (0-360°), and the azimuth range is the direction of the GNSS mobile station (centered on the mobile station azimuth, 40° as the range) to obtain two sets of snow depth inversion results. In the present invention, in particular, a set of snow depth inversion results is obtained for the under-snow receiver range. Since only the under-snow receiver can obtain SWE, we obtain the snow depth SD in this range to calculate the density. Utilizing the stable property of regional density, it can be multiplied by the SD representing a larger range to obtain the SWE representing a larger range.

[0042] Finally, it is necessary to process two groups of SD and one group of SWE: the SD in a larger range is obtained in all directions, and the SD in the direction of the mobile station and one group of SWE have a matching area, which is used to solve the density of the intermediate quantity.

[0043] S3. Execute the GNSS refraction SWE inversion algorithm in batches to solve the relative positioning baseline between the GNSS base station and the GNSS mobile station. The sampling interval of the GNSS base station observation file ensures that it matches the time of the GNSS mobile station file. Apply the correction file and estimate the floating point solution of the L1+L2+L5 triple frequency using the observation equation and ionosphere-free combination through combined filtering. The starting plane is the location of the mobile antenna under the snow.

[0044] S4. Data cleaning and quality check. Perform data cleaning and quality check for the calculated SWE and two sets of snow depths.

[0045] S5. Calculate the regional snow density for the snow depth and SWE within the same azimuth range. Calculate the regional snow density ρ = SWE / SD using the snow depth and SWE within the same azimuth range.

[0046] S6. The SWE of the extended range is obtained by multiplying the regional snow density by the snow depth in the full angular range. The density does not change much in a small area, less than 10%, as an intermediate quantity, and the snow depth representing the full angular range is again converted into SWE to expand the monitoring range of SWE.

[0047] Example 1

[0048] Figure 1It is a set of vertically assembled GNSS receiver equipment, including a GNSS mobile station receiver, a GNSS mobile station antenna, a GNSS base station receiver and a GNSS base station antenna. Among them, the GNSS base station and mobile station receiver are used to generate and store GNSS observation files and GNSS broadcast ephemeris files. The GNSS base station antenna is placed on the mast and is used to execute the GNSS interferometric snow depth inversion algorithm to obtain the snow depth. The GNSS mobile station antenna is buried under the snow and is used to calculate SWE. In this embodiment, the experimental site is located in Dronning Maud Land, Antarctica. The GNSS mobile station antenna was located at an azimuth of 221° from the GNSS base station antenna. The GNSS mobile station antenna and the GNSS base station antenna were fixed relative to each other and the mast, and all measurements were made relative to each other. The sampling intervals of the two receivers were different, 1 second for the GNSS base station and 30 seconds for the mobile station.

[0049] like Figure 2 As shown, the regional snow water equivalent inversion method based on the vertical combined GNSS receiver of the present invention includes the following steps:

[0050] S1. Reduce the sampling frequency of the base station receiver and download resource files. The sampling interval of the GNSS base station receiver located on the mast is reduced to 15s. Download the IGS precise ephemeris, IGS precise clock and antenna correction files for each station for the day for the two stations.

[0051] Specifically: Use GFZRNX to resample the GNSS base station observation files to 15 seconds, because the sampling interval has little effect on the GNSS interferometric snow depth inversion algorithm. The two stations downloaded the resource files for GNSS relative positioning, including the final precise ephemeris and clock error files provided by IGS, and the corresponding antenna correction files, of which igs14.atx before doy331 in 2022 and igs20.atx after.

[0052] S2. Execute the GNSS interferometric snow depth inversion algorithm. The selected signal frequencies include S1C, S2S and S2Q, excluding S2W. The preset elevation angle range is 5°-30°. The signal distance limit range in the LSP diagram is 0-8 meters. The daily execution azimuth range is no mask (0-360°) and the azimuth range is the direction of the mobile station (centered on the mobile station azimuth, 40° as the range) to obtain two sets of snow depth inversion results.

[0053] Specifically, in this embodiment, the noisiest S2W frequency among the GNSS signal frequencies received by the GNSS base station is excluded, the elevation angle range is set to 5°-30°, the maximum distance is limited to 8 meters, and the GNSS interferometric snow depth algorithm is executed for no azimuth mask (0-360°) and azimuth mask (201°-241°) respectively. The basic principle of the algorithm is as follows:

[0054] The inversion of the GNSS interferometric snow depth algorithm uses the phase difference between the reflected signal and the direct signal to reflect the plane height, i.e., the snow depth. The signal-to-noise ratio (SNR) data used is expressed as the signal strength received by the GNSS antenna. In the low altitude angle range, the SNR is specifically the direct signal power (P d ) and reflected signal power (P r ) between them.

[0055]

[0056] Where φ is the phase difference between the two signals and the SNR in dB is usually converted to a linear scale (volts). When the direct signal is removed from the fit using a low-order polynomial, the cosine reflected signal dependence on the satellite elevation angle E can be obtained.

[0057]

[0058] Among them, S r represents the reflected signal sequence, A is the amplitude, λ is the wavelength of the carrier phase observation, is the phase shift.

[0059] The GNSS interferometric snow depth algorithm uses an adaptive Lomb-Scargle periodogram (LSP) to obtain the main frequency f=2h / λ, thereby obtaining the reflection height h. Since the satellite orbit in the site reflection area may be affected by building occlusion, tree diffraction and terrain undulation, abnormal results may occur. Further quality control is required to set parameter thresholds (peak signal-to-noise ratio is 3, primary and secondary signal-to-noise ratio is 1.5). The final SD obtained by the algorithm is the difference between the reflection height and the plane reference surface. In this embodiment, the plane reference surface is corrected by field data.

[0060] S3. Execute the GNSS refraction SWE inversion algorithm in batches to solve the relative positioning baseline of the GNSS base station and the GNSS mobile station. The sampling interval of the GNSS base station observation file is to ensure the time match with the GNSS mobile station file. Apply the correction file and estimate the floating point solution of the L1+L2+L5 triple frequency using the observation equation and ionosphere-free combination through combined filtering. The starting plane is the location of the mobile antenna under the snow.

[0061] Specifically: The basic principle of the GNSS refraction SWE inversion algorithm is as follows:

[0062] The carrier phase observations recorded by the receiver under the snow are affected by the liquid water in the snow, which reduces the signal propagation speed in the snow and produces signal delay. As the air-snow interface medium changes, the GNSS signal is refracted, resulting in a change in the angle of incidence into the receiver. The additional error source δL caused by the snow layer m It is defined as the excess path length between the receiver below the snowpack and the receiver above the snowpack when positioned relative to each other, as follows:

[0063]

[0064] Where λ is the wavelength of the carrier phase observation; is the phase observation value, which is different from the Satellite pair {k,l}; is the normalized line-of-sight vector between the satellite and receiver antenna phase centers; is the baseline vector between the rod and the ground antenna; all known path delays (from the antenna under the snow to the GPS satellite), including atmospheric delay, integer ambiguity, multipath error and other errors; (·) kl =(·) k -(·) l is the difference between satellites k and l.

[0065] When the baseline solution is performed for the above and below snowpack receivers without removing this error term, the Up component in the fixed baseline result will change. The SWE is derived from the deviation of the Up component, and the change is shown as follows:

[0066] δSWE≈δUp (4)

[0067] Although there are other interfering errors such as atmospheric effects, these errors appear only as Gaussian white noise within very short baselines and the impact is less than 1 cm. For combined receivers within a few meters, the most accurate estimate is obtained using carrier phase observations at all elevation angles without elevation-dependent weighting.

[0068] The relative positioning baseline of the GNSS base station and the GNSS mobile station is solved using RTKLIB b34. The sampling interval of the GNSS base station observation file is 1 second to ensure time matching with the mobile station file with a time interval of 30 seconds. The full elevation angle is used, as well as the IGS precise ephemeris, IGS precise clock and antenna correction files for each station (igs20.atx after 2022 doy331). The floating point solution of the L1+L2+L5 triple frequency is estimated using the observation equations and the ionosphere-free combination through combined filtering. The starting plane is the location of the mobile antenna under the snow, and the height difference between the two antennas is 325 cm, which becomes 625 cm after 2022 doy39.

[0069] S4. Perform data cleaning to perform quality checks on the calculated SWE and the two sets of snow depths.

[0070] Specifically: count the valid days of snow depth and SWE obtained under two azimuths, exclude the days with different data, and exclude the days with too large standard deviation of snow depth within the azimuth.

[0071] S5. Calculate the regional snow density for the snow depth and SWE within the same azimuth range.

[0072] Specifically: use the density conversion formula ρ = SWE / SD, where SD is the snow depth.

[0073] S6. Use the product of regional snow density and snow depth in all angular ranges to obtain the SWE of the extended range.

[0074] Specifically, the conversion formula in step S5 is used to obtain SWE again. Since this SWE is calculated from the snow depth representing the full angular range, it can represent a larger range in meaning. The same density is used as the intermediate quantity for conversion because the density does not change much in a small area, less than 10%.

[0075] like Figure 3 As shown in Figure 2, the evaluation and comparison of the effect of the method include the representative in situ SWE results, the SWE results estimated by the density model, the regional scale SWE results obtained by the method of the invention, and the reference SWE. The method of the invention shows a similar pattern with several other results, and has a higher determination coefficient (R 2 >0.948), proving the stability of the algorithm. However, unlike the SWE results that represent the original site, it provides a more representative SWE result with a larger monitoring range. The uncertainty of the SWE obtained by the method of the present invention is further evaluated, using the error propagation law, starting from the posterior results of snow depth, snow density, and SWE verification with field reference values, including the following formula:

[0076]

[0077] Among them, σ ρ is the uncertainty of the in-situ density, SWE is the value obtained by the GNSS refraction SWE inversion algorithm, SD_GNSS_Azi is the snow depth obtained by the GNSS interferometric snow depth inversion algorithm under the azimuth mask, σ SD_GNSS_Azi is the uncertainty of the snow depth obtained by the GNSS interferometric snow depth inversion algorithm under the azimuth mask. In this embodiment, it is half of the root mean square error of the different orbit results obtained every day. SWE is the uncertainty of SWE obtained by GNSS refraction SWE inversion algorithm, which is 40 mm in this embodiment. Similarly, SD_GNSS_Aziall is the snow depth obtained by GNSS interferometric snow depth inversion algorithm at all angles, σ SD_GNSS_Azi The uncertainty of snow depth is obtained by GNSS interferometric snow depth inversion algorithm at all angles.

[0078] The calculated a posteriori error σ SWE_region The range is between 41.32 mm and 193.72 mm, with about 65% below 60 mm (~10%). Although these uncertainties exceed the uncertainty of manual measurement (40 mm or ~10%), they are still within 1.5 times the uncertainty of manual SWE. The uncertainty of the verified extended range SWE is 60 mm (~10%).

[0079] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A regional snow water equivalent inversion method based on a vertically combined GNSS receiver, characterized in that: The following steps are involved: S1. Base station receiver sampling frequency downsampling and resource file downloading: The sampling interval of the GNSS base station receiver is downsampled to 15s, and the IGS precise ephemeris, IGS precise clock and antenna correction files for each station corresponding to the day are downloaded for the two stations; S2. Execute the GNSS interferometric snow depth inversion algorithm. The selected signal frequencies include S1C, S2S and S2Q, excluding S2W. The preset elevation angle range is 5°-30°. The limit range of the signal distance in the LSP diagram is 0 to 8 meters. The daily execution azimuth range is 0-360° without setting a mask. The azimuth range is the direction of the GNSS mobile station. The azimuth angle is centered at the mobile station azimuth and the range is 40°. Two sets of snow depth inversion results are obtained, and one set of snow depth inversion results is obtained for the receiver range under the snow; S3. Batch execute the GNSS refraction SWE inversion algorithm to solve the relative positioning baseline between the GNSS base station and the GNSS mobile station. The sampling interval of the GNSS base station observation file ensures that the time matches the GNSS mobile station file; apply the correction file, use the observation equation and the ionosphere-free combination to estimate the floating point solution of the L1+L2+L5 triple frequency through combined filtering, and the starting plane is the location of the mobile antenna under the snow; S4, data cleaning and quality check: perform data cleaning and quality check for the calculated SWE and two sets of snow depths; S5, calculating the regional snow density for the snow depth and SWE within the same azimuth range, and calculating the regional snow density ρ = SWE / SD using the snow depth and SWE within the same azimuth range; S6. Use the product of regional snow density and snow depth in all-round angular range to obtain SWE in the extended range. The density does not change much in a small area, less than 10%, as an intermediate quantity. The snow depth representing the all-round angular range is again converted into SWE to expand the monitoring range of SWE.

2. The regional snow water equivalent inversion method based on a vertically combined GNSS receiver according to claim 1 is characterized in that: Step S1, the GNSS base station receiver is located on the mast, and the GNSS mobile station receiver is buried underground at a position of 221° azimuth of the GNSS base station. The sampling interval of the GNSS base station receiver is 1 second, and the sampling interval of the GNSS mobile station receiver is 30 seconds.

3. The regional snow water equivalent inversion method based on a vertically combined GNSS receiver according to claim 1 is characterized in that: Step S2 is specifically as follows: excluding the noisiest S2W frequency among the GNSS signal frequencies received by the GNSS base station, setting the altitude angle range to 5°-30°, limiting the maximum distance to 8 meters, and executing the GNSS interferometric snow depth algorithm for the case where no azimuth mask is set, i.e. 0-360°, and the case where an azimuth mask is set, i.e. 201°-241°.

4. The regional snow water equivalent inversion method based on a vertically combined GNSS receiver according to claim 3 is characterized in that: Step S2, the inversion of the GNSS interferometric snow depth algorithm uses the phase difference between the reflected signal and the direct signal to reflect the plane height, i.e., the snow depth. The signal-to-noise ratio (SNR) data used is expressed as the signal strength received by the GNSS antenna. In the low altitude angle range, the SNR is specifically the direct signal power P d and the reflected signal power P r Interference between: Where φ is the phase difference between the two signals. The unit dB of SNR is usually converted to a linear scale. When the direct signal is removed from the fit using a low-order polynomial, the cosine reflected signal dependence on the satellite elevation angle E is obtained: Among them, S r represents the reflected signal sequence, A is the amplitude, θ is the carrier wavelength, is the phase shift; The main frequency f = 2h / λ is obtained using the adaptive Lomb-Scargle periodogram LSP, thereby obtaining the reflection height h; the parameter threshold is set, the peak signal-to-noise ratio is 3, and the primary and secondary signal-to-noise ratio is 1.

5. The final SD obtained is the difference between the reflection height and the plane reference surface.

5. The regional snow water equivalent inversion method based on a vertically combined GNSS receiver according to claim 1 is characterized in that: Step S3 is specifically: Additional error source δL caused by snow layer m Defined as the excess path length between the receiver under the snowpack and the receiver above the snowpack when positioned relative to each other: Where λ is the wavelength of the carrier phase; is the phase observation value; satellite pair {k,l}; is the normalized line-of-sight vector between the satellite and receiver antenna phase centers; is the baseline vector between the rod and the ground antenna; all known path delays (from the antenna under the snow to the GPS satellite), including atmospheric delay, integer ambiguity, multipath error and other errors. (·) kl =(·) k -(·) l is the difference between satellites k and l; When the baseline solution is performed for the receivers above and below the snowpack without removing the error term, the Up component in the fixed baseline result will change. The SWE is derived from the deviation of the Up component, and the change is shown in the following formula: δSWE≈δUp (4) RTKLIB b34 is used to solve the relative positioning baseline of the GNSS reference station and the GNSS mobile station; the sampling interval of the reference station observation file is 1 second, using the full elevation angle, as well as the IGS precise ephemeris, IGS precise clock and antenna correction files for each station. The floating-point solution of the L1+L2+L5 triple frequency is estimated by combined filtering using the observation equation and ionosphere free combination. The starting plane is the location of the mobile antenna under the snow. The height difference between the two antennas is 325 cm, which becomes 625 cm after doy39 in 2022.

6. The regional snow water equivalent inversion method based on a vertically combined GNSS receiver according to claim 1 is characterized in that: Step S4 is specifically as follows: counting the valid days of snow depth and SWE obtained under two azimuths, excluding days with different data, and excluding days with excessively large standard deviation of snow depth within the azimuth.

Images