Snow depth inversion method and system based on combined satellite remote sensing and spaceborne GNSS-R technology

Abstract

The present invention discloses a snow depth inversion method and system combining satellite remote sensing and satellite-borne GNSS-R technology, the method comprising: obtaining the shallow snow area of ​​the study area in a long time series based on satellite remote sensing technology; obtaining the surface reflectivity of the shallow snow area in the same time series based on satellite-borne GNSS-R technology; inputting the surface reflectivity of the shallow snow area and the specified surface characteristic parameters into a trained neural network model, and outputting the snow depth of the shallow snow area. The present invention utilizes satellite-borne GNSS-R technology and satellite remote sensing technology, combined with a neural network model to estimate snow depth, thereby improving the accuracy of snow depth estimation.

Description

Technical Field

[0001] The present invention belongs to the field of snow depth inversion, and specifically relates to a snow depth inversion method and system combining satellite remote sensing and spaceborne GNSS-R technology. Background Art

[0002] Snow covers play an important role in global climate change and water cycle, and monitoring the temporal and spatial changes of snow covers is of great significance. There are two main methods for snow cover monitoring: ground station measurement and satellite remote sensing monitoring.

[0003] Traditional ground station measurements are to obtain snow depth at fixed stations and use interpolation methods to fit the discrete points to large-scale surface data. However, the disadvantage of this method is that the ground stations represent a small range and the stations are unevenly distributed.

[0004] Satellite remote sensing monitoring of snow accumulation mainly includes optical remote sensing and passive microwave remote sensing. Among them, optical remote sensing data can only extract the snow coverage area; microwave remote sensing observation data can invert and extract the snow depth. However, although passive microwave remote sensing has strong penetration and all-weather working characteristics, it has low spatial resolution, and the missed judgment rate and misjudgment rate are high in areas with shallow snow depth (less than 10cm). Summary of the invention

[0005] In order to overcome the shortcomings of the above-mentioned prior art, the present invention provides a snow depth inversion method and system combining satellite remote sensing and space-borne GNSS-R technology, which utilizes space-borne GNSS-R technology and satellite remote sensing technology in combination with a neural network model to estimate snow depth and improve the accuracy of snow depth estimation.

[0006] According to one aspect of the present invention, a snow depth inversion method combining satellite remote sensing and spaceborne GNSS-R technology is provided, comprising:

[0007] Obtain shallow snow areas in the study area within a long time series based on satellite remote sensing technology;

[0008] The surface reflectivity of shallow snow areas in the same time series is obtained based on the spaceborne GNSS-R technology;

[0009] The surface reflectivity of the shallow snow area and the specified surface characteristic parameters are input into the trained neural network model, and the snow depth of the shallow snow area is output; wherein the training of the neural network model includes:

[0010] Obtain shallow snow areas in the study area within a long time series based on satellite remote sensing technology;

[0011] The surface reflectivity of shallow snow areas in the same time series is obtained based on the spaceborne GNSS-R technology;

[0012] Obtain snow depth data in shallow snow areas within the same time series based on ground station measurements;

[0013] Matching the reflectivity of the shallow snow area and the specified surface characteristic parameters with the snow depth according to time and location to obtain a training data set;

[0014] Constructing a neural network model, wherein the input layer of the neural network model is the surface characteristic parameters of the specified area at different times, and the output layer is the snow depth of the corresponding area;

[0015] The constructed neural network model is trained using the training data set, and the trained neural network model is output.

[0016] As a further technical solution, the specified surface characteristic parameters include soil temperature and soil moisture.

[0017] As a further technical solution, the shallow snow area in the study area within a long time series is obtained based on satellite remote sensing technology, including:

[0018] Use optical remote sensing satellites to obtain snow cover data in the study area over a long time series;

[0019] Passive microwave remote sensing satellites are used to obtain snow depth data in the same time series and snow-covered areas;

[0020] The snow depth data is used to screen out shallow snow areas where the snow depth is lower than a preset value.

[0021] As a further technical solution, the method also includes: constructing a verification data set, inputting the verification data set into the trained neural network model, and obtaining a snow depth product inverted by the satellite-borne GNSS-R.

[0022] As a further technical solution, the method also includes: fusing the snow depth product inverted by the satellite-borne GNSS-R with the snow depth product obtained by passive microwave remote sensing to obtain a comprehensive snow depth product.

[0023] According to one aspect of the present invention, there is provided a snow depth inversion system combining satellite remote sensing and spaceborne GNSS-R technology, comprising:

[0024] The first main module is used to obtain the shallow snow area in the study area within a long time series based on satellite remote sensing technology;

[0025] The second main module is used to obtain the surface reflectivity of the shallow snow area in the same time series based on the spaceborne GNSS-R technology;

[0026] The third main module is used to input the surface reflectivity of the shallow snow area and the specified surface characteristic parameters into the trained neural network model, and output the snow depth of the shallow snow area; wherein the training of the neural network model includes:

[0027] Obtain shallow snow areas in the study area within a long time series based on satellite remote sensing technology;

[0028] The surface reflectivity of shallow snow areas in the same time series is obtained based on the spaceborne GNSS-R technology;

[0029] Obtain snow depth data in shallow snow areas within the same time series based on ground station measurements;

[0030] Matching the reflectivity of the shallow snow area and the specified surface characteristic parameters with the snow depth according to time and location to obtain a training data set;

[0031] Constructing a neural network model, wherein the input layer of the neural network model is the surface characteristic parameters of the specified area at different times, and the output layer is the snow depth of the corresponding area;

[0032] The constructed neural network model is trained using the training data set, and the trained neural network model is output.

[0033] As a further technical solution, the first main module further includes:

[0034] The first submodule is used to obtain snow cover data in the study area in a long time series using optical remote sensing satellites;

[0035] The second submodule is used to obtain snow depth data in the same time series and snow-covered area using passive microwave remote sensing satellites;

[0036] The third submodule is used to use the snow depth data to filter out shallow snow areas where the snow depth is lower than a preset value.

[0037] According to one aspect of the present invention specification, there is provided an electronic device, comprising a processor and a memory, wherein the memory stores program instructions executed by the processor, and the processor calls the program instructions to execute the steps of the snow depth inversion method combining satellite remote sensing and spaceborne GNSS-R technology.

[0038] According to one aspect of the present specification, there is provided a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions enable the computer to perform the steps of the snow depth inversion method combining satellite remote sensing and spaceborne GNSS-R technology.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The present invention proposes to combine the spaceborne GNSS-R technology with two currently commonly used satellite remote sensing technologies, and eliminate the deep snow area range of passive microwave remote sensing from the snow cover area range of optical remote sensing, so as to obtain the spaceborne GNSS-R snow inversion range; and utilize the sensitivity of the spaceborne GNSS-R technology to the snow depth, and use the surface reflectivity (SR) parameters obtained by the spaceborne GNSS-R technology in the shallow snow area in combination with the surface characteristic parameters to construct a snow depth estimation model; in order to express the nonlinear relationship between multiple surface characteristic parameters and snow depth, a neural network (NN for short) is used to construct the snow depth estimation model. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, a brief introduction is given below to the drawings used in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0042] Figure 1 A schematic flow chart of a snow depth inversion method combining satellite remote sensing and spaceborne GNSS-R technology provided in an embodiment of the present invention.

[0043] Figure 2 A schematic diagram of the structure of a snow depth inversion system combining satellite remote sensing and spaceborne GNSS-R technology provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0044] It should be noted that:

[0045] Spaceborne GNSS-R technology uses GNSS to measure the delay (time delay or phase delay) between the direct signal and the signal reflected by the surface mirror, and inverts surface reflectivity, soil moisture and other surface characteristics based on the geometric position relationship between the GNSS satellite, receiver and mirror reflection point. The surface reflectivity-difference ratio factor (SR-DR factor) provided by spaceborne GNSS-R products is highly correlated with snow depth and can be used to extract snow depth.

[0046] The terms "including" and "having" and any variations thereof in the specification and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusions, for example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to the steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or apparatuses.

[0047] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. In addition, the technical features in the various embodiments or single embodiments provided by the present invention are arbitrarily combined with each other to form a new technical solution. This combination is not restricted by the sequence of steps and / or the structural composition mode, but must be based on the ability of ordinary technicians in this field to achieve. When the combination of technical solutions is contradictory or cannot be achieved, it should be considered that this combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0048] The embodiment of the present invention provides a snow depth inversion method combining satellite remote sensing and spaceborne GNSS-R technology, such as Figure 1 As shown, including:

[0049] Step 1: Obtain the shallow snow area in the study area within a long time series based on satellite remote sensing technology.

[0050] Step 2: Obtain the surface reflectivity of the shallow snow area in the same time series based on the spaceborne GNSS-R technology.

[0051] Step 3, inputting the surface reflectivity of the shallow snow area and the specified surface characteristic parameters into the trained neural network model, and outputting the snow depth in the shallow snow area.

[0052] Wherein, the training of the neural network model includes:

[0053] Step 3.1, obtain the shallow snow area in the study area within the long time series based on satellite remote sensing technology.

[0054] Specifically, optical remote sensing satellites are used to obtain snow cover data in the study area over a long time series. Passive microwave remote sensing satellites are used to obtain snow depth data in the snow-covered area over the same time series, and areas where the snow depth is lower than a preset value (such as 10 cm) are screened out.

[0055] The spaceborne GNSS-R technology is used to obtain the surface reflectivity of the shallow snow area in the same time series, and other surface characteristic parameters (soil temperature and soil moisture) are obtained.

[0056] Step 3.2, obtain high-precision snow depth data of shallow snow areas measured by ground stations in the same time series.

[0057] Step 3.3, matching the reflectivity of the shallow snow area and the specified surface characteristic parameters with the snow depth according to time and location to obtain a training data set.

[0058] Specifically, the above-mentioned multiple surface characteristic parameters (soil temperature, soil moisture, surface reflectivity) and snow depth values ​​in the shallow snow area are matched according to time and location, and the dataset is divided into a training set and a validation set.

[0059] Step 3.4, construct a multi-layer neural network model, the input layer of the neural network model is the surface characteristic parameters of the specified area at different times, and the output layer is the snow depth of the corresponding area.

[0060] Step 3.5, using the training data set to train the constructed neural network model, and outputting the trained neural network model.

[0061] Specifically, the training set and the validation set are input into the multi-layer neural network model for training, and the model parameters are saved.

[0062] After obtaining the trained multi-layer neural network model, the method further includes: performing snow depth inversion based on the multi-layer neural network model. Inputting the surface characteristic parameters of the non-training data set into the constructed multi-layer neural network model to obtain the snow depth product inverted by the satellite-borne GNSS-R.

[0063] Furthermore, the snow depth product inverted by the space-borne GNSS-R is integrated with the snow depth product obtained by passive microwave remote sensing to obtain a comprehensive snow depth product.

[0064] As a preferred embodiment, the embodiment of the present invention provides an embodiment of snow depth estimation modeling based on a back propagation neural network using FY-3C satellite products and space-borne GNSS-R products, which specifically includes the following steps:

[0065] (1) Download the snow cover data provided by the FY-3C satellite snow cover product (MULSS_SNC) for the period from October 2023 to October 2024.

[0066] (2) Download the snow depth data and snow water equivalent product (MWRIX_SWE) provided by the FY-3C satellite microwave imager in the same time series and extract the snow cover area, and filter out the areas where the snow depth is less than 10 cm.

[0067] (3) Download the snow depth data of the shallow snow area extracted in the same time series from the National Meteorological Science Data Sharing Service Platform (http: / / data.cma.cn / site / index.html).

[0068] (4) Download the surface reflectivity data provided by the spaceborne GNSS-R products in shallow snow areas within the same time series, the soil temperature data of the European Centre for Medium-Range Weather Forecasts (ECMWF) Atmospheric Reanalysis Dataset v5 (ERA5), and the global daily 36 km EASE grid soil moisture data of the Soil Moisture Active and Passive (SMAP) L3 radiometer.

[0069] (5) Multiple surface characteristic parameters (soil temperature, soil moisture, surface reflectivity) and snow depth values ​​in the above-mentioned shallow snow area are aligned and matched according to time and location, and the dataset is randomly divided into a training set and a validation set (here the dataset ratio is set to 7:3).

[0070] (6) Construct a back propagation neural network (BPNN) model with 3 hidden layers. The input layer is the surface characteristic parameters of the above-mentioned specified area at different times, and the output layer is the corresponding snow depth.

[0071] (7) Input the training set and validation set data into BPNN, train the model until convergence, and save the model parameters.

[0072] (8) Snow depth inversion based on BPNN. The surface characteristic parameters of the area to be inverted are input into the BPNN model to obtain the snow depth product inverted by the satellite-borne GNSS-R.

[0073] (9) The snow depth product inverted by the satellite GNSS-R is integrated with the snow depth product obtained by passive microwave remote sensing to obtain a comprehensive snow depth product.

[0074] The implementation basis of each embodiment of the present invention is to implement programmed processing through a device with a processor function. Therefore, in engineering practice, the technical solutions and functions of each embodiment of the present invention are encapsulated into various modules. Based on this reality, on the basis of the above embodiments, an embodiment of the present invention provides a snow depth inversion system combining satellite remote sensing and space-borne GNSS-R technology, which is used to execute the snow depth inversion method combining satellite remote sensing and space-borne GNSS-R technology in the above method embodiment.

[0075] See also Figure 2The system includes: a first main module, which is used to obtain the shallow snow area of ​​the study area within a long time series based on satellite remote sensing technology; a second main module, which is used to obtain the surface reflectivity of the shallow snow area within the same time series based on the satellite-borne GNSS-R technology; a third main module, which is used to input the surface reflectivity of the shallow snow area and the specified surface characteristic parameters into the trained neural network model, and output the snow depth of the shallow snow area; wherein the training of the neural network model includes: obtaining the shallow snow area of ​​the study area within a long time series based on satellite remote sensing technology; obtaining the surface reflectivity of the shallow snow area within the same time series based on the satellite-borne GNSS-R technology; obtaining the snow depth data of the shallow snow area within the same time series based on ground station measurement; matching the reflectivity of the shallow snow area and the specified surface characteristic parameters with the snow depth according to time and position to obtain a training data set; constructing a neural network model, the input layer of the neural network model is the surface characteristic parameters of the specified area at different times, and the output layer is the snow depth of the corresponding area; using the training data set to train the constructed neural network model, and output the trained neural network model.

[0076] The snow depth inversion system provided by the embodiment of the present invention combines satellite remote sensing and spaceborne GNSS-R technology to solve the problems existing in existing snow monitoring. Figure 2 Several modules in the system use space-borne GNSS-R technology and satellite remote sensing technology, combined with a neural network model to estimate snow depth and improve the accuracy of snow depth estimation.

[0077] It should be noted that the system embodiment provided by the present invention is used to implement the methods in the above method embodiment as well as the methods in other method embodiments provided by the present invention. The only difference is that the corresponding functional modules are set. The principle is basically the same as the principle of the above system embodiment provided by the present invention. As long as the technical personnel in this field refer to the specific technical solutions in other method embodiments on the basis of the above system embodiment, obtain the corresponding technical means and the technical solutions composed of these technical means by combining the technical features, and on the premise of ensuring the practicality of the technical solutions, improve the modules in the above system embodiment to obtain the corresponding system class embodiments, which are used to implement the methods in other method class embodiments. For example:

[0078] Based on the content of the above system embodiment, as a preferred embodiment, the snow depth inversion system combining satellite remote sensing and spaceborne GNSS-R technology provided in the embodiment of the present invention, the first main module also includes:

[0079] The first submodule is used to obtain snow cover data in the study area in a long time series using optical remote sensing satellites;

[0080] The second submodule is used to obtain snow depth data in the same time series and snow-covered area using passive microwave remote sensing satellites;

[0081] The third submodule is used to use the snow depth data to filter out shallow snow areas where the snow depth is lower than a preset value.

[0082] Based on the contents of the above system embodiments, as a preferred embodiment, in the snow depth inversion system combining satellite remote sensing and spaceborne GNSS-R technology provided in the embodiments of the present invention, the specified surface characteristic parameters include soil temperature and soil moisture.

[0083] Based on the content of the above system embodiment, as a preferred embodiment, in the snow depth inversion system of combined satellite remote sensing and spaceborne GNSS-R technology provided in the embodiment of the present invention, the first main module is also used to execute the following instructions:

[0084] Use optical remote sensing satellites to obtain snow cover data in the study area over a long time series;

[0085] Passive microwave remote sensing satellites are used to obtain snow depth data in the same time series and snow-covered areas;

[0086] The snow depth data is used to screen out shallow snow areas where the snow depth is lower than a preset value.

[0087] Based on the content of the above system embodiment, as a preferred embodiment, the snow depth inversion system of the combined satellite remote sensing and spaceborne GNSS-R technology provided in the embodiment of the present invention further includes:

[0088] The fourth main module is used to construct a verification data set, input the verification data set into the trained neural network model, and obtain the snow depth product inverted by the space-borne GNSS-R.

[0089] Based on the content of the above system embodiment, as a preferred embodiment, the snow depth inversion system of the combined satellite remote sensing and spaceborne GNSS-R technology provided in the embodiment of the present invention further includes:

[0090] The fifth main module is used to merge the snow depth product inverted by the space-borne GNSS-R with the snow depth product obtained by passive microwave remote sensing to obtain a comprehensive snow depth product.

[0091] An embodiment of the present invention further provides an electronic device, including a processor and a memory, wherein the memory stores program instructions executed by the processor, and the processor calls the program instructions to execute the steps of the snow depth inversion method of combining satellite remote sensing and spaceborne GNSS-R technology, including:

[0092] Obtain shallow snow areas in the study area within a long time series based on satellite remote sensing technology;

[0093] The surface reflectivity of shallow snow areas in the same time series is obtained based on the spaceborne GNSS-R technology;

[0094] The surface reflectivity of the shallow snow area and the specified surface characteristic parameters are input into the trained neural network model, and the snow depth of the shallow snow area is output; wherein the training of the neural network model includes:

[0095] Obtain shallow snow areas in the study area within a long time series based on satellite remote sensing technology;

[0096] The surface reflectivity of shallow snow areas in the same time series is obtained based on the spaceborne GNSS-R technology;

[0097] Obtain snow depth data in shallow snow areas within the same time series based on ground station measurements;

[0098] Matching the reflectivity of the shallow snow area and the specified surface characteristic parameters with the snow depth according to time and location to obtain a training data set;

[0099] Constructing a neural network model, wherein the input layer of the neural network model is the surface characteristic parameters of the specified area at different times, and the output layer is the snow depth of the corresponding area;

[0100] The constructed neural network model is trained using the training data set, and the trained neural network model is output.

[0101] The embodiment of the present invention further provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, wherein the computer instructions enable the computer to perform the steps of the snow depth inversion method combining satellite remote sensing and spaceborne GNSS-R technology, including:

[0102] Obtain shallow snow areas in the study area within a long time series based on satellite remote sensing technology;

[0103] The surface reflectivity of shallow snow areas in the same time series is obtained based on the spaceborne GNSS-R technology;

[0104] The surface reflectivity of the shallow snow area and the specified surface characteristic parameters are input into the trained neural network model, and the snow depth of the shallow snow area is output; wherein the training of the neural network model includes:

[0105] Obtain shallow snow areas in the study area within a long time series based on satellite remote sensing technology;

[0106] The surface reflectivity of shallow snow areas in the same time series is obtained based on the spaceborne GNSS-R technology;

[0107] Obtain snow depth data in shallow snow areas within the same time series based on ground station measurements;

[0108] Matching the reflectivity of the shallow snow area and the specified surface characteristic parameters with the snow depth according to time and location to obtain a training data set;

[0109] Constructing a neural network model, wherein the input layer of the neural network model is the surface characteristic parameters of the specified area at different times, and the output layer is the snow depth of the corresponding area;

[0110] The constructed neural network model is trained using the training data set, and the trained neural network model is output.

[0111] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A snow depth inversion method combining satellite remote sensing and spaceborne GNSS-R technology, characterized in that: include: Obtain shallow snow areas in the study area within a long time series based on satellite remote sensing technology; The surface reflectivity of shallow snow areas in the same time series is obtained based on the spaceborne GNSS-R technology; The surface reflectivity of the shallow snow area and the specified surface characteristic parameters are input into the trained neural network model, and the snow depth of the shallow snow area is output; wherein the training of the neural network model includes: Obtain shallow snow areas in the study area within a long time series based on satellite remote sensing technology; The surface reflectivity of shallow snow areas in the same time series is obtained based on the spaceborne GNSS-R technology; Obtain snow depth data in shallow snow areas within the same time series based on ground station measurements; Matching the reflectivity of the shallow snow area and the specified surface characteristic parameters with the snow depth according to time and location to obtain a training data set, wherein the specified surface characteristic parameters include soil temperature and soil moisture; Constructing a neural network model, wherein the input layer of the neural network model is the surface characteristic parameters of the specified area at different times, and the output layer is the snow depth of the corresponding area; The constructed neural network model is trained using the training data set, and the trained neural network model is output.

2. The snow depth inversion method combining satellite remote sensing and spaceborne GNSS-R technology according to claim 1 is characterized in that: The shallow snow area in the study area within a long time series is obtained based on satellite remote sensing technology, including: Use optical remote sensing satellites to obtain snow cover data in the study area over a long time series; Passive microwave remote sensing satellites are used to obtain snow depth data in the same time series and snow-covered areas; The snow depth data is used to screen out shallow snow areas where the snow depth is lower than a preset value.

3. The snow depth inversion method combining satellite remote sensing and spaceborne GNSS-R technology according to claim 1 is characterized in that: The method also includes: constructing a verification data set, inputting the verification data set into a trained neural network model, and obtaining a snow depth product inverted by a satellite-borne GNSS-R.

4. The snow depth inversion method combining satellite remote sensing and spaceborne GNSS-R technology according to claim 3 is characterized in that: The method also includes: fusing the snow depth product inverted by the satellite-borne GNSS-R with the snow depth product obtained by passive microwave remote sensing to obtain a comprehensive snow depth product.

5. A snow depth inversion system combining satellite remote sensing and spaceborne GNSS-R technology, characterized by: include: The first main module is used to obtain the shallow snow area in the study area within a long time series based on satellite remote sensing technology; The second main module is used to obtain the surface reflectivity of the shallow snow area in the same time series based on the spaceborne GNSS-R technology; The third main module is used to input the surface reflectivity of the shallow snow area and the specified surface characteristic parameters into the trained neural network model, and output the snow depth of the shallow snow area; wherein the training of the neural network model includes: Obtain shallow snow areas in the study area within a long time series based on satellite remote sensing technology; The surface reflectivity of shallow snow areas in the same time series is obtained based on the spaceborne GNSS-R technology; Obtain snow depth data in shallow snow areas within the same time series based on ground station measurements; Matching the reflectivity of the shallow snow area and the specified surface characteristic parameters with the snow depth according to time and location to obtain a training data set, wherein the specified surface characteristic parameters include soil temperature and soil moisture; Constructing a neural network model, wherein the input layer of the neural network model is the surface characteristic parameters of the specified area at different times, and the output layer is the snow depth of the corresponding area; The constructed neural network model is trained using the training data set, and the trained neural network model is output.

6. The snow depth inversion system combining satellite remote sensing and spaceborne GNSS-R technology according to claim 5 is characterized in that: The first main module also includes: The first submodule is used to obtain snow cover data in the study area in a long time series using optical remote sensing satellites; The second submodule is used to obtain snow depth data in the same time series and snow-covered area using passive microwave remote sensing satellites; The third submodule is used to use the snow depth data to filter out shallow snow areas where the snow depth is lower than a preset value.

7. An electronic device, characterized in that: It includes a processor and a memory, the memory stores program instructions executed by the processor, and the processor calls the program instructions to execute the steps of the snow depth inversion method of combining satellite remote sensing and space-borne GNSS-R technology as described in any one of claims 1 to 4.

8. A non-transitory computer readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, which enable the computer to execute the steps of the snow depth inversion method combining satellite remote sensing and space-borne GNSS-R technology as described in any one of claims 1 to 4.

Images