A method and system for lidar satellite inversion of background aerosol optical parameters

Abstract

This invention provides a method and system for satellite-based lidar inversion of background aerosol optical parameters, comprising: preprocessing spaceborne lidar calibration data to obtain vertical channel, parallel channel, and hyperspectral channel signals of background aerosols; employing different denoising strategies to invert daytime and nighttime backscattering coefficients of the vertical channel, parallel channel, and hyperspectral channel signals to obtain nighttime backscattering coefficients and final backscattering coefficients; performing noise reduction processing on the cumulative atmospheric optical thickness corresponding to the nighttime signal to obtain the nighttime extinction coefficient; constructing a lidar-specific climate field from the nighttime backscattering coefficient and the nighttime extinction coefficient; and multiplying the lidar-specific climate field with the final backscattering coefficient to obtain the final extinction coefficient. This invention effectively overcomes the challenge of low signal-to-noise ratio and significantly improves inversion accuracy by introducing innovative multi-scale denoising and systematic bias correction techniques.

Description

Technical Field

[0001] This invention relates to the field of remote sensing meteorological inversion technology, and in particular to a lidar satellite inversion method and system for background aerosol optical parameters. Background Technology

[0002] Aerosols exert a significant influence on the Earth's climate system through direct scattering and absorption of solar radiation (direct effects) and by influencing cloud properties and lifetimes as cloud condensation nuclei (indirect effects). Among these, background aerosols, which are widely distributed but have weak optical signals, play a crucial role in global radiation balance and aerosol-cloud interactions, despite often being undetected by conventional detection algorithms. Studies have shown that aerosols not detected by CALIPSO (Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations) layer detection algorithms can contribute approximately 20% of aerosol optical thickness. Satellite lidar, as an active remote sensing tool, can provide three-dimensional vertical distribution information of aerosols and clouds, making it a key means of studying background aerosols.

[0003] Current satellite remote sensing methods are divided into two categories: active remote sensing and passive remote sensing. Commonly used passive detection methods, such as the Moderate-resolution Imaging Spectroradiometer (MODIS), can only receive observation signals of atmospheric column integration and cannot obtain information on the vertical distribution of clouds and aerosols. In active remote sensing, spaceborne lidar detects clouds and aerosols by actively emitting lasers. It has the advantages of high spatiotemporal resolution, high accuracy, and all-weather observation, making it one of the main tools for detecting the vertical distribution of clouds and aerosols. The Aerosol and Carbon Dioxide Detection Lidar (ACDL) is the first satellite detection system to employ high spectral resolution cloud and aerosol detection technology, possessing enormous potential for high-precision cloud and aerosol detection. At a wavelength of 532 nm, ACDL uses an iodine absorption filter to separate the Mie scattering of aerosols and the Rayleigh scattering of atmospheric molecules. Therefore, it can independently invert the aerosol extinction coefficient and backscattering coefficient without assuming lidar ratios, providing an unprecedented data source for the accurate detection of background aerosols. However, when ACDL detects background aerosols, its signal-to-noise ratio is extremely low, especially during the day, which severely limits the effective use of its data.

[0004] Currently, representative existing technologies for the extraction and inversion of weak aerosol signals mainly include:

[0005] Option 1: CALIPSO Level 3 Stratospheric Aerosol Profile Product. This option enhances the detection capability of stratospheric background aerosols by increasing the calibration altitude to 30-40 km and performing monthly averaging of data within a 5° latitudinal × 20° meridional grid. Its main feature is sacrificing spatial and temporal resolution for a lower detection limit, making it suitable for studying large-scale, long-term averaged stratospheric aerosol distributions.

[0006] Option 2: CALIPSO Inversion Method Based on SAGE III / ISS Solar Occultation Data Constraints. This method utilizes high-precision extinction coefficient profile data from SAGE III / ISS as constraints to invert transient aerosol extinction coefficients in the CALIPSO Level 1B signal that were not identified by the hierarchical detection algorithm. This method extends the nighttime detection limit to... The magnitude is large, but due to noise interference during the day, the inversion uncertainty is as high as 454.5%, and its feasibility depends on the synchronous observation of external satellite data (SAGE III / ISS).

[0007] Understandably, the aforementioned methods have several drawbacks: the spatial resolution of CALIPSO Level 3 products is coarse, and its 5°×20° monthly average grid cannot capture transient events (such as the initial diffusion process of a volcanic eruption), and the monthly average product cannot meet the requirements of near-real-time applications. Furthermore, the CALIOP iso-miel scattering lidar signal is ill-conditioned, requiring the assumption of a fixed lidar ratio when retrieving the extinction coefficient, introducing uncertainties exceeding 100% in high-latitude regions. The SAGE III / ISS-constrained CALIPSO inversion method is significantly affected by solar background noise during the day, resulting in substantial inversion uncertainty and limiting its practicality. Moreover, this method relies on external data and must be matched with SAGE III / ISS observations, limiting its independence and widespread application. Existing ACDL standard data processing algorithms typically focus on strong aerosol and cloud signals identified by hierarchical detection algorithms, while ignoring or struggling to effectively process weak aerosol signals in the background. Summary of the Invention

[0008] This invention provides a method and system for inverting the optical parameters of background aerosols using a lidar satellite, addressing the shortcomings of existing technologies. It enables high-quality inversion of the optical properties of global three-dimensional background aerosols from raw observation signals acquired by a spaceborne high spectral resolution lidar (HSRL), including aerosol backscattering coefficients, extinction coefficients, and the crucial lidar ratio. The method fully considers the significant differences between day and night signals in the actual observation environment of spaceborne lidar, the statistical characteristics of noise, and the sensitivity of different inversion steps to noise. Through a series of interconnected and refined processing steps, a high-precision product is ultimately generated.

[0009] In a first aspect, the present invention provides a method for lidar satellite inversion of background aerosol optical parameters, comprising:

[0010] Preprocessing of spaceborne lidar calibration data yields vertical channel, parallel channel, and hyperspectral channel signals of background aerosols.

[0011] Different denoising strategies were employed to invert the day and night backscattering coefficients of the vertical channel signal, parallel channel signal, and hyperspectral channel signal, resulting in the nighttime backscattering coefficient and the final backscattering coefficient.

[0012] The cumulative atmospheric optical thickness corresponding to the nighttime signal is denoised to obtain the nighttime extinction coefficient;

[0013] The lidar climate field is constructed using the nighttime backscattering coefficient and the nighttime extinction coefficient.

[0014] The final extinction coefficient is obtained by multiplying the climate field by the lidar and the final backscattering coefficient.

[0015] According to the present invention, a method for satellite inversion of background aerosol optical parameters for lidar is provided, which preprocesses the calibration data of the onboard lidar to obtain the vertical channel signal, parallel channel signal, and hyperspectral channel signal of the background aerosol, including:

[0016] A hierarchical detection algorithm is used to process the calibration data of the spaceborne lidar, identify and remove signals from strong aerosol layers, cloud layers and attenuation areas below clouds, and retain the vertical channel signal, parallel channel signal and hyperspectral channel signal of background aerosols.

[0017] According to the lidar satellite inversion method for background aerosol optical parameters provided by the present invention, different denoising strategies are employed to invert the day and night backscattering coefficients of the vertical channel signal, the parallel channel signal, and the hyperspectral channel signal to obtain the final backscattering coefficients, including:

[0018] Daytime and nighttime signals were extracted from the vertical channel signal, parallel channel signal, and hyperspectral channel signal, respectively.

[0019] High-frequency noise is eliminated from the daytime signal using a preset wavelet transform, and the processed daytime and nighttime signals are filtered using a two-dimensional mean filter to obtain the first daytime signal and the first nighttime signal.

[0020] The ratio of the vertical channel signal to the parallel channel signal in the first daytime signal and the first nighttime signal is used as the first ratio, and the ratio of the parallel channel signal to the hyperspectral channel signal in the first daytime signal and the first nighttime signal is used as the second ratio.

[0021] The data is divided into several window blocks according to a fixed scale. The ratio of the mean of the parallel channel signal to the mean of the hyperspectral channel signal in each window block is used as the reference value of the current window block. The median corresponding to the second ratio in the current window block is calculated. The offset is determined by the reference value and the median. The second ratio is corrected based on the offset and the preset window to obtain the corrected second ratio.

[0022] The final backscattering coefficients are generated by inversion based on the first ratio and the corrected second ratio.

[0023] According to the present invention, a lidar satellite inversion method for background aerosol optical parameters employs different denoising strategies to invert day and night backscattering coefficients of vertical channel signals, parallel channel signals, and hyperspectral channel signals, obtaining nighttime backscattering coefficients, including:

[0024] Extract the nighttime signal from the vertical channel signal, parallel channel signal, and hyperspectral channel signal;

[0025] The night signal is subjected to high-frequency noise processing using a preset wavelet transform, and the processed night signal is filtered using SG filtering to obtain the second night signal.

[0026] The ratio of the vertical channel signal to the parallel channel signal in the second night signal is used as the third ratio, and the ratio of the parallel channel signal to the hyperspectral channel signal in the second night signal is used as the fourth ratio.

[0027] The data is divided into several window blocks according to a fixed scale. The ratio of the mean of the parallel channel signal to the mean of the hyperspectral channel signal in each window block is used as the reference value of the current window block. The median corresponding to the fourth ratio in the current window block is calculated. The offset is determined by the reference value and the median. The fourth ratio is corrected based on the offset and the preset window to obtain the corrected fourth ratio.

[0028] Inversion is performed based on the third ratio and the corrected fourth ratio to generate the nighttime backscattering coefficient.

[0029] According to the lidar satellite inversion method for background aerosol optical parameters provided by the present invention, the cumulative atmospheric optical thickness corresponding to the nighttime signal is denoised to obtain the nighttime extinction coefficient, including:

[0030] The cumulative atmospheric optical thickness is obtained by integrating the aerosol and molecular extinction coefficients along the laser path in the second night signal.

[0031] The cumulative atmospheric optical thickness was filtered using SG filtering to obtain the nighttime extinction coefficient.

[0032] According to the present invention, a lidar satellite inversion method for background aerosol optical parameters is provided, which constructs a lidar-specific climate field from the nighttime backscattering coefficient and the nighttime extinction coefficient, including:

[0033] The ratio of aerosol lidar is obtained by inverting the ratio of nighttime extinction coefficient to nighttime backscattering coefficient;

[0034] The instantaneous value of the aerosol lidar ratio is calculated based on the nighttime inversion results, and a lidar ratio climate field is constructed.

[0035] According to the present invention, a lidar satellite inversion method for background aerosol optical parameters is provided, which calculates the instantaneous value of the aerosol lidar ratio based on the nighttime inversion results and constructs a lidar ratio climate field, including:

[0036] The global area is divided into a preset three-dimensional grid, and the effective aerosol lidar ratio within the preset physical range is aggregated into the corresponding location grid.

[0037] All signals are iterated over, and an average grid is generated based on the cumulative array through element-wise division, with empty arrays counted as missing values.

[0038] Traverse the missing values, use a breadth-first search algorithm to obtain one or more values ​​that are closest to the missing value, and take the average value to fill the missing unit;

[0039] A 3D Gaussian filter with a preset small standard deviation is used to smooth the small-scale noise of the aerosol lidar ratio, thus forming a lidar ratio climate field.

[0040] In a second aspect, the present invention also provides a lidar satellite inversion system for background aerosol optical parameters, comprising:

[0041] The preprocessing module is used to preprocess the calibration data of the spaceborne lidar to obtain the vertical channel signal, parallel channel signal and hyperspectral channel signal of the background aerosol;

[0042] The first inversion module is used to invert the day and night backscattering coefficients of the vertical channel signal, parallel channel signal and hyperspectral channel signal using different denoising strategies, so as to obtain the nighttime backscattering coefficient and the final backscattering coefficient.

[0043] The second inversion module is used to perform noise reduction processing on the cumulative atmospheric optical thickness corresponding to the nighttime signal to obtain the nighttime extinction coefficient.

[0044] The third inversion module is used to construct the lidar-ratio climate field from the nighttime backscattering coefficient and the nighttime extinction coefficient;

[0045] The fourth inversion module is used to obtain the final extinction coefficient by multiplying the climate field by the lidar and the final backscattering coefficient.

[0046] Thirdly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement a lidar satellite inversion method for background aerosol optical parameters as described above.

[0047] Fourthly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a lidar satellite inversion method for background aerosol optical parameters as described above.

[0048] The lidar satellite inversion method and system for background aerosol optical parameters provided by this invention have the following beneficial effects:

[0049] (1) Strong independence: It does not rely on external data sources such as SAGE, and can complete the inversion of the entire set of optical parameters using only ACDL's own observation data, which significantly improves its independent application capability;

[0050] (2) High accuracy and small bias: Addressing the signal-to-noise ratio difference between day and night data, multi-scale denoising and innovative systematic bias correction effectively suppressed noise influence, significantly reducing the proportion of negative values ​​and systematic bias in the inversion results. The extinction coefficient obtained from the inversion showed high consistency with the SAGE III / ISS extinction product. The mean absolute error (MAE) for nighttime and daytime data was 6.06 × 10⁻⁶. and 7.71× ;

[0051] (3) High spatiotemporal resolution: It can provide instantaneous, high-resolution three-dimensional optical parameters and capture transient events such as the diffusion process of sulfate aerosols after volcanic eruptions;

[0052] (4) The first assumption-free inversion of the global three-dimensional background aerosol lidar ratio was achieved: overcoming the limitation of traditional methods that require assumptions about the lidar ratio, and providing more reliable key parameters for aerosol type identification and climate effect research. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0054] Figure 1 This is one of the flowcharts of the lidar satellite inversion method for background aerosol optical parameters provided by the present invention;

[0055] Figure 2 This is the second schematic diagram of the process for the lidar satellite inversion method for background aerosol optical parameters provided by the present invention;

[0056] Figure 3 This is a schematic diagram of the structure of the lidar satellite inversion system for background aerosol optical parameters provided by the present invention;

[0057] Figure 4 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0059] The background aerosol has an extremely low optical thickness, and its signal is extremely weak, resulting in a severely insufficient signal-to-noise ratio and the following drawbacks:

[0060] Weak signal and high noise interference: especially during the day, the solar background noise is 1-2 orders of magnitude higher than at night, which makes the original signal submerged by noise and the traditional inversion method fail.

[0061] There is a systematic bias: When inverting the backscattering coefficient, the noise in the hyperspectral channel signal has a specific distribution. As a divisor, it will introduce a systematic negative bias into the ratio of the attenuated backscattering coefficients of the parallel channel and the hyperspectral channel. This will cause the inverted aerosol backscattering coefficient and extinction coefficient to be seriously underestimated or even have a large number of physically meaningless negative values.

[0062] The significant differences between day and night signals make it difficult to process them uniformly: the significant differences in signal-to-noise ratio between day and night require differentiated processing strategies, but existing methods lack targeted and effective solutions that can simultaneously guarantee the quality and consistency of day and night data.

[0063] The lidar ratio with high spatiotemporal resolution cannot be directly obtained: Traditional Mie scattering lidar (such as CALIPSO) requires the assumption of lidar ratio. Even when using data from my country's spaceborne high spectral resolution lidar ACDL, due to the aforementioned noise and bias issues, it is difficult to stably and accurately inverse the lidar ratio from weak instantaneous signals, let alone construct its global three-dimensional distribution.

[0064] The aforementioned issues prevent existing technologies from reliably acquiring global, three-dimensional, high-resolution background aerosol optical properties (backscattering coefficient, extinction coefficient, lidar ratio) from spaceborne HSRL data such as ACDL, thus limiting their application in accurately quantifying aerosol radiative forcing and studying aerosol-cloud interactions.

[0065] To address the shortcomings of existing technologies, this invention proposes a method for inverting background aerosol optical parameters based on a spaceborne hyperspectral resolution lidar (HSRL). A complete, independent data processing workflow capable of effectively handling extremely low signal-to-noise ratio challenges is constructed. The aim is to invert the optical properties of global three-dimensional background aerosols, including aerosol backscattering coefficients, extinction coefficients, and the crucial lidar ratio, from the raw observation signals acquired by the spaceborne HSRL. The entire scheme fully considers the significant differences between day and night signals in the actual observation environment of the spaceborne lidar, the statistical characteristics of noise, and the sensitivity of different inversion steps to noise. Through interconnected and refined processing, high-precision products are ultimately generated.

[0066] The technical terms used in this invention are explained as follows:

[0067] Aerosols are colloidal dispersion systems formed by the dispersion and suspension of solid or liquid small particles in a gaseous medium. They are also known as gaseous dispersion systems, and their particle size is approximately 0.001–100 μm.

[0068] Background aerosols: These are weak aerosols with extremely low optical thickness that are usually not effectively identified by the standard layer detection algorithm of spaceborne lidar, but which are persistent and widespread in the atmosphere.

[0069] Optical thickness: The integral value of the extinction coefficient along the laser transmission path, which reflects the overall attenuation of the laser signal by the atmosphere and is an important parameter of atmospheric radiation transmission.

[0070] High-spectral-resolution lidar (HSRL): A type of lidar that can separate aerosol Mie scattering signals from atmospheric molecule Rayleigh scattering signals through spectral filtering techniques (such as iodine absorption cells), and can independently retrieve aerosol extinction coefficients and backscattering coefficients.

[0071] Extinction coefficient: The relative attenuation rate of electromagnetic radiation as it propagates through the atmosphere due to particle absorption and scattering; it is the sum of the scattering coefficient and the absorption coefficient.

[0072] Backscattering coefficient: refers to the scattering coefficient corresponding to the scattering of waves, particles or signals reflected back to their source direction.

[0073] Attenuated backscattering coefficient: The backscattering coefficient that includes the effect of atmospheric transmission attenuation is a physical quantity that is directly obtained after calibration of the original signal received by the lidar.

[0074] LiDAR ratio: The ratio of aerosol extinction coefficient to its backscattering coefficient (unit: Sr) is a key parameter characterizing aerosol type and optical properties.

[0075] Signal-to-noise ratio (SNR): The ratio of effective signal strength to background noise strength, which is an important indicator for measuring the detection capability of lidar.

[0076] Climatological field: refers to a dataset that is statistically derived from long-term observation data and characterizes the average spatiotemporal distribution of a certain physical quantity (such as lidar ratio).

[0077] Figure 1 This is one of the flowcharts illustrating the lidar satellite inversion method for background aerosol optical parameters provided in this embodiment of the invention, such as... Figure 1 As shown, it includes:

[0078] Step 100: Preprocess the calibration data of the spaceborne lidar to obtain the vertical channel signal, parallel channel signal and hyperspectral channel signal of the background aerosol;

[0079] Step 200: Using different denoising strategies, the day and night backscattering coefficients of the vertical channel signal, parallel channel signal and hyperspectral channel signal are inverted to obtain the nighttime backscattering coefficient and the final backscattering coefficient;

[0080] Step 300: Perform noise reduction processing on the cumulative atmospheric optical thickness corresponding to the nighttime signal to obtain the nighttime extinction coefficient;

[0081] Step 400: Construct the lidar-based climate field using the nighttime backscattering coefficient and the nighttime extinction coefficient;

[0082] Step 500: The final extinction coefficient is obtained by multiplying the climate field by the lidar and the final backscattering coefficient.

[0083] Specifically, the inversion process of background aerosol optical parameters of the spaceborne lidar in this embodiment of the invention is as follows: Figure 2 As shown, it includes:

[0084] The first step is signal preprocessing:

[0085] A corresponding hierarchical detection algorithm is applied to the calibration data of the spaceborne lidar to identify and remove signals from strong aerosol layers, cloud layers, and attenuated regions below clouds. Only the three-channel signals of background aerosols are retained. , and Its expression is as follows:

[0086] (1)

[0087] (2)

[0088] (3)

[0089] in, β ( r The detection range is... r The backscattering coefficient at 532 nm, where the superscripts ⊥, ∥, and M represent the vertical, parallel, and hyperspectral channels, respectively. For the detection range is r Vertical channel backscattering coefficient at that time For the detection range is r Parallel channel backscattering coefficient at that time For the detection range is r Hyperspectral channel backscattering coefficient at that time

[0090] The subscripts m and a represent atmospheric molecules and aerosols, respectively. For the detection range is r Atmospheric molecules in the vertical channel at that time For the detection range is r Vertical channel aerosol at that time For the detection range is r Parallel channel atmospheric molecules at that time For the detection range is r Parallel channel aerosols at that time Indicates the detection range as r Molecular transmittance of the iodine molecule filter Indicates aerosol permeability. The detection range is r Atmospheric bidirectional transmittance related to optical thickness τ (i.e., the integral of extinction coefficient α).

[0091] The second step is the backscattering coefficient inversion, which yields the final backscattering coefficient and the nighttime backscattering coefficient:

[0092] Different denoising strategies were employed to retrieve the backscattering coefficients for the three-channel signals during both day and night. Nighttime backscattering coefficient retrieval used a two-dimensional mean filter with a kernel size of a certain scale (e.g., 55 km horizontally × 0.504 km vertically). Daytime retrieval first used a typical wavelet basis function sym3 or other wavelet basis with a certain number of decomposition levels (e.g., 6) to eliminate high-frequency noise, followed by the same mean filtering process as nighttime retrieval. Figure 2 The kernel size used in the mean filtering is 1. Due to the influence of solar cosmic rays, some daytime signals exhibit vertically continuous salt-and-pepper noise bands. These noise bands can be detected by gradient analysis. The detection method is that if the average gradient value of a profile exceeds the average gradient value of all profiles in the scene, the profile is considered to be contaminated by salt-and-pepper noise, and the affected areas above 20 km altitude are replaced by two-dimensional interpolation.

[0093] It should be noted that the embodiments of the present invention may also employ different wavelet basis functions or other time-frequency analysis tools, such as empirical mode decomposition (EMD) to replace the sym3 wavelet for signal-noise separation.

[0094] Furthermore, for ease of illustration, the embodiments of the present invention introduce... and Representing atmospheric depolarization ratio and and The ratio:

[0095] (4)

[0096] (5)

[0097] The calculation is affected by noise. The existence of a systematic negative bias leads to an underestimation of the calculated backscattering and extinction coefficients. To address this issue, this invention proposes a method to correct this by calculating the average value of the background aerosol signal within the block. The core idea of ​​this method is that when there are enough sampling points, the deviation in the ratio of the average values ​​of two signals can be ignored. The specific processing flow is as follows: First, the data is divided into several blocks according to a fixed scale; then, within each window block, the parallel channel signal is calculated. ) and hyperspectral channel signals ( The ratio of the means is used as the baseline value for this block; then the median of the K values ​​corresponding to all denoised spectral units within the window block is calculated, and the offset E between this median and the aforementioned reference value is calculated:

[0098] (6)

[0099] in, To take the median function, Parallel channel signal Take the average. Hyperspectral channel signal Take the average.

[0100] Finally, from and The matrix formed by the ratio Subtract E from the middle to complete the deviation correction.

[0101] In the above correction process, an appropriate window size needs to be determined to balance the accuracy of bias estimation and spatial resolution. This invention employs a systematic window convergence analysis method to determine a suitable window size. Given the diurnal differences in signal-to-noise ratio and potential bias magnitude, the embodiment uses a fixed window size (W×H) for bias correction at night and during the day (e.g., 1 pixel (24 m) vertically at night and 21 pixels (504 m) during the day, and 50 pixels (250 km) horizontally both day and night). Other window sizes (W×H) can be used for bias correction in this case.

[0102] Using the corrected ratio and known molecular parameters, the aerosol backscattering coefficient is obtained by inversion using formula (7), and finally, aerosol backscattering coefficient products for both day and night are generated:

[0103] (7)

[0104] in This is the depolarization ratio of atmospheric molecules, which is calculated to be 0.00366 according to atmospheric models. For the detection range is r Depolarization ratio at time, atmospheric molecular extinction coefficient and molecular backscattering coefficient The transmittance of iodine molecular filters can be calculated based on Rayleigh scattering theory. and aerosol transmittance For details of the calculation method, please refer to the relevant papers. In this embodiment, the temperature and pressure data used to calculate the above parameters are taken from the Modern Reanalysis Research Applications Edition 2 (MERRA-2) reanalysis dataset, but other similar datasets can also be used.

[0105] The third step is the nighttime extinction coefficient inversion:

[0106] Because extinction coefficient inversion is extremely sensitive to noise, large-scale noise reduction processing of relevant optical parameters is required during lidar ratio inversion. The backscattering coefficient and extinction coefficient used for lidar ratio inversion are obtained through the following integrated process: A two-dimensional wavelet transform with a certain number of decomposition levels (e.g., 6) combined with a large-scale Savitzky-Golay (SG) filter (e.g., 55 km × 1.992 km) is used to reduce noise in the attenuated backscattered signal. The cumulative atmospheric optical thickness is then calculated based on the corrected ratio and the denoised signal. The cumulative optical thickness is the integral of the aerosol and molecular extinction coefficients along the laser path, and can be obtained using equations (1) to (5):

[0107] (8)

[0108] in, For the laser path, For detection distance r The integral variable, The aerosol integral component of the extinction coefficient. The atmospheric molecular integral component of the extinction coefficient. For the detection range is r Atmospheric molecules at that time For the detection range is r Atmospheric bidirectional transmittance and atmospheric molecular components at that time For the detection range is r Atmospheric bidirectional transmittance aerosol component at that time.

[0109] After smoothing the atmospheric optical thickness using an SG filter of the same size as the attenuated backscattered signal, for example... Figure 2 The kernel size of the SG filter is The preliminary nighttime aerosol extinction coefficient is derived using equation (9):

[0110] (9)

[0111] The fourth step is to construct the lidar ratio climate field:

[0112] aerosol lidar ratio It is a key parameter for classifying aerosol types, determined by the absorption and scattering characteristics of atmospheric particulate matter, and can be obtained by inverting the ratio of the extinction coefficient to the backscattering coefficient as described above:

[0113] (10)

[0114] Next, based on the nighttime inversion results, instantaneous lidar ratios are calculated to construct a global climate field for all four seasons. The globe is divided into grids, for example, 1°×1°×1 km (longitude×latitude×altitude). When constructing the lidar ratio climate field, the grid resolution can be adjusted to other scales as needed. Effective lidar ratios within a reasonable physical range (e.g., 0~100 sr) are aggregated into the corresponding grid locations. After signal traversal, an average grid is generated based on the cumulative array using element-wise division, and empty arrays are counted as missing values. Subsequently, the missing values ​​are traversed, and a breadth-first search (BFS) algorithm is used to find one or more values ​​closest to the missing value, and their average is used to fill the missing cells. Finally, a 3D Gaussian filter with a small standard deviation (e.g., 1 grid) is used to smooth the small-scale noise of the lidar ratios, and the final output is the global distribution of the lidar ratios for all four seasons.

[0115] The fifth step is the inversion of the final extinction coefficient:

[0116] By multiplying the backscattering coefficient obtained by inversion with the ratio of the meteorological lidar at the corresponding spatiotemporal location, the extinction coefficient is finally obtained, and the aerosol extinction coefficient product is output.

[0117] The background aerosol optical parameter inversion system provided by the present invention is described below. The background aerosol optical parameter inversion system described below can be referred to in correspondence with the background aerosol optical parameter inversion method described above.

[0118] Figure 3 This is a schematic diagram of the structure of the lidar satellite inversion system for background aerosol optical parameters provided in an embodiment of the present invention, as shown below. Figure 3 As shown, it includes: a preprocessing module 31, a first inversion module 32, a second inversion module 33, a third inversion module 34, and a fourth inversion module 35, wherein:

[0119] The preprocessing module 31 is used to preprocess the calibration data of the spaceborne lidar to obtain the vertical channel signal, parallel channel signal and hyperspectral channel signal of the background aerosol; the first inversion module 32 is used to invert the day and night backscattering coefficients of the vertical channel signal, parallel channel signal and hyperspectral channel signal using different denoising strategies to obtain the nighttime backscattering coefficient and the final backscattering coefficient; the second inversion module 33 is used to denoise the cumulative atmospheric optical thickness corresponding to the nighttime signal to obtain the nighttime extinction coefficient; the third inversion module 34 is used to construct the lidar relative climate field from the nighttime backscattering coefficient and the nighttime extinction coefficient; the fourth inversion module 35 is used to multiply the lidar relative climate field with the final backscattering coefficient to obtain the final extinction coefficient.

[0120] Figure 4An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4 As shown, the electronic device may include a processor 410, a communication interface 420, a memory 430, and a communication bus 440. The processor 410, communication interface 420, and memory 430 communicate with each other via the communication bus 440. The processor 410 can call logical instructions in the memory 430 to execute a lidar satellite inversion method for background aerosol optical parameters. This method includes: preprocessing the onboard lidar calibration data to obtain the vertical channel signal, parallel channel signal, and hyperspectral channel signal of the background aerosol; employing different denoising strategies to invert the day and night backscattering coefficients of the vertical channel signal, parallel channel signal, and hyperspectral channel signal to obtain the nighttime backscattering coefficient and the final backscattering coefficient; performing denoising processing on the cumulative atmospheric optical thickness corresponding to the nighttime signal to obtain the nighttime extinction coefficient; constructing a lidar-specific climate field from the nighttime backscattering coefficient and the nighttime extinction coefficient; and multiplying the lidar-specific climate field with the final backscattering coefficient to obtain the final extinction coefficient.

[0121] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0122] On the other hand, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, this computer program implements a lidar satellite inversion method for background aerosol optical parameters provided by the methods described above. This method includes: preprocessing spaceborne lidar calibration data to obtain vertical channel signals, parallel channel signals, and hyperspectral channel signals of background aerosols; employing different denoising strategies to invert day and night backscattering coefficients of the vertical channel signals, parallel channel signals, and hyperspectral channel signals to obtain nighttime backscattering coefficients and final backscattering coefficients; performing noise reduction processing on the cumulative atmospheric optical thickness corresponding to the nighttime signals to obtain nighttime extinction coefficients; constructing a lidar-specific climate field from the nighttime backscattering coefficients and nighttime extinction coefficients; and multiplying the lidar-specific climate field by the final backscattering coefficients to obtain the final extinction coefficient. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment, depending on actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0123] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0124] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for lidar satellite inversion of background aerosol optical parameters, characterized in that, include: Preprocessing of spaceborne lidar calibration data yields vertical channel, parallel channel, and hyperspectral channel signals of background aerosols. Different denoising strategies were employed to invert the day and night backscattering coefficients of the vertical channel signal, parallel channel signal, and hyperspectral channel signal, resulting in the nighttime backscattering coefficient and the final backscattering coefficient. The cumulative atmospheric optical thickness corresponding to the nighttime signal is denoised to obtain the nighttime extinction coefficient; The lidar climate field is constructed using the nighttime backscattering coefficient and the nighttime extinction coefficient. The final extinction coefficient is obtained by multiplying the climate field by the lidar and the final backscattering coefficient.

2. The method for lidar satellite inversion of background aerosol optical parameters according to claim 1, characterized in that, Preprocessing of spaceborne lidar calibration data yields the vertical channel signal, parallel channel signal, and hyperspectral channel signal of background aerosols, including: A hierarchical detection algorithm is used to process the calibration data of the spaceborne lidar, identify and remove signals from strong aerosol layers, cloud layers and attenuation areas below clouds, and retain the vertical channel signal, parallel channel signal and hyperspectral channel signal of background aerosols.

3. The method for lidar satellite inversion of background aerosol optical parameters according to claim 1, characterized in that, Different denoising strategies were employed to invert the day and night backscattering coefficients of the vertical channel signal, parallel channel signal, and hyperspectral channel signal, yielding the final backscattering coefficients, including: Daytime and nighttime signals were extracted from the vertical channel signal, parallel channel signal, and hyperspectral channel signal, respectively. High-frequency noise is eliminated from the daytime signal using a preset wavelet transform, and the processed daytime and nighttime signals are filtered using a two-dimensional mean filter to obtain the first daytime signal and the first nighttime signal. The ratio of the vertical channel signal to the parallel channel signal in the first daytime signal and the first nighttime signal is used as the first ratio, and the ratio of the parallel channel signal to the hyperspectral channel signal in the first daytime signal and the first nighttime signal is used as the second ratio. The data is divided into several window blocks according to a fixed scale. The ratio of the mean of the parallel channel signal to the mean of the hyperspectral channel signal in each window block is used as the reference value of the current window block. The median corresponding to the second ratio in the current window block is calculated. The offset is determined by the reference value and the median. The second ratio is corrected based on the offset and the preset window to obtain the corrected second ratio. The final backscattering coefficients are generated by inversion based on the first ratio and the corrected second ratio.

4. The method for lidar satellite inversion of background aerosol optical parameters according to claim 1, characterized in that, Different denoising strategies were employed to invert the daytime and nighttime backscattering coefficients of the vertical channel, parallel channel, and hyperspectral channel signals, yielding the nighttime backscattering coefficients, including: Extract the nighttime signal from the vertical channel signal, parallel channel signal, and hyperspectral channel signal; The night signal is subjected to high-frequency noise processing using a preset wavelet transform, and the processed night signal is filtered using SG filtering to obtain the second night signal. The ratio of the vertical channel signal to the parallel channel signal in the second night signal is used as the third ratio, and the ratio of the parallel channel signal to the hyperspectral channel signal in the second night signal is used as the fourth ratio. The data is divided into several window blocks according to a fixed scale. The ratio of the mean of the parallel channel signal to the mean of the hyperspectral channel signal in each window block is used as the reference value of the current window block. The median corresponding to the fourth ratio in the current window block is calculated. The offset is determined by the reference value and the median. The fourth ratio is corrected based on the offset and the preset window to obtain the corrected fourth ratio. Inversion is performed based on the third ratio and the corrected fourth ratio to generate the nighttime backscattering coefficient.

5. The lidar satellite inversion method for background aerosol optical parameters according to claim 4, characterized in that, The cumulative atmospheric optical thickness corresponding to the nighttime signal is denoised to obtain the nighttime extinction coefficient, including: The cumulative atmospheric optical thickness is obtained by integrating the aerosol and molecular extinction coefficients along the laser path in the second night signal. The cumulative atmospheric optical thickness was filtered using SG filtering to obtain the nighttime extinction coefficient.

6. The method for lidar satellite inversion of background aerosol optical parameters according to claim 1, characterized in that, The lidar-based climate field is constructed using the nighttime backscattering coefficient and the nighttime extinction coefficient, including: The ratio of aerosol lidar is obtained by inverting the ratio of nighttime extinction coefficient to nighttime backscattering coefficient; The instantaneous value of the aerosol lidar ratio is calculated based on the nighttime inversion results, and a lidar ratio climate field is constructed.

7. The lidar satellite inversion method for background aerosol optical parameters according to claim 6, characterized in that, Based on the nighttime inversion results, the instantaneous value of the aerosol lidar ratio is calculated, and a lidar ratio climate field is constructed, including: The global area is divided into a preset three-dimensional grid, and the effective aerosol lidar ratio within the preset physical range is aggregated into the corresponding location grid. All signals are iterated over, and an average grid is generated based on the cumulative array through element-wise division, with empty arrays counted as missing values. Traverse the missing values, use a breadth-first search algorithm to obtain one or more values ​​that are closest to the missing value, and take the average value to fill the missing unit; A 3D Gaussian filter with a preset small standard deviation is used to smooth the small-scale noise of the aerosol lidar ratio, thus forming a lidar ratio climate field.

8. A lidar satellite inversion system for background aerosol optical parameters, characterized in that, include: The preprocessing module is used to preprocess the calibration data of the spaceborne lidar to obtain the vertical channel signal, parallel channel signal and hyperspectral channel signal of the background aerosol; The first inversion module is used to invert the day and night backscattering coefficients of the vertical channel signal, parallel channel signal and hyperspectral channel signal using different denoising strategies, so as to obtain the nighttime backscattering coefficient and the final backscattering coefficient. The second inversion module is used to perform noise reduction processing on the cumulative atmospheric optical thickness corresponding to the nighttime signal to obtain the nighttime extinction coefficient. The third inversion module is used to construct the lidar-ratio climate field from the nighttime backscattering coefficient and the nighttime extinction coefficient; The fourth inversion module is used to obtain the final extinction coefficient by multiplying the climate field by the lidar and the final backscattering coefficient.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the lidar satellite inversion method for background aerosol optical parameters as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the lidar satellite inversion method for background aerosol optical parameters as described in any one of claims 1 to 7.

Images