A method, device and medium for quantitatively evaluating spatial ozone distribution in sand-dust passage

Abstract

This application discloses a method, equipment, and medium for quantitative assessment of spatial ozone distribution during dust storms, relating to the fields of atmospheric environmental chemistry and remote sensing monitoring. The method includes: unifying the time reference and resampling the spatiotemporal grid of the two-dimensional observation matrix representing dust and the two-dimensional observation matrix representing ozone; determining the ozone anomaly matrix based on the ozone observation matrix and typical clean days; decomposing the dust observation matrix and the ozone anomaly matrix into multiple sequence pairs according to altitude; performing sliding correlation analysis along the time axis for each sequence pair, calculating the Pearson correlation coefficient, and determining the optimal time offset for that altitude layer; extracting time-aligned sequence pairs under the optimal time offset and calculating the Spearman rank correlation coefficient; and stitching the optimal time offsets output from all altitude layers along the altitude axis to generate a continuous vertical profile, and combining this with the Spearman rank correlation coefficient to output a composite layer, significantly improving the statistical robustness and physical accuracy of the quantitative assessment results.

Description

Technical Field

[0001] This application relates to the field of atmospheric environmental chemistry and remote sensing monitoring, and in particular to a method, equipment and medium for quantitative assessment of spatial ozone distribution during dust storms. Background Technology

[0002] In the field of atmospheric environmental chemistry and remote sensing monitoring, the passage of dust clouds not only brings particulate matter pollution, but also exerts a complex dual photochemical effect on ozone formation and depletion by altering radiation flux and providing heterogeneous reaction interfaces. To achieve quantitative assessment of spatial ozone distribution during dust storms, related technologies still face the following three challenges in practical data processing and mechanism analysis: 1. Spatiotemporal resolution mismatch of multi-source heterogeneous matrices: Traditional studies often rely on single-dimensional concentration observations near the ground, making it difficult to reveal the drastically different coupling mechanisms at higher altitudes (e.g., 0-3000 m). When introducing vertical observation techniques, the two-dimensional observation matrix for dust characterization (e.g., gridded extinction coefficient data retrieved from lidar) and the two-dimensional observation matrix for ozone characterization may originate from different instruments. These different instruments exhibit significant differences in sampling frequency and vertical spatial resolution.

[0003] 2. Conventional background noise masks the true physical disturbance characteristics: Ozone concentration exhibits strong, naturally occurring diurnal variations. Existing time-series comparison methods often directly use observed absolute values ​​for calculation, failing to effectively remove the highly dependent static background baseline. This unprocessed background noise severely interferes with sliding correlation calculations, preventing algorithms from accurately extracting anomalous ozone increases or decreases purely caused by dust disturbances.

[0004] 3. Single sliding algorithms struggle to balance temporal accuracy and nonlinear physical response: The evolutionary characteristics of dust storms and ozone not only exhibit temporal misalignments ("leading" or "lagging"), but their coupling strength often displays complex nonlinearity. Traditional techniques for quantifying this coupling effect rely solely on a single cross-correlation algorithm, which has inherent statistical limitations: on the one hand, conventional algorithms struggle to avoid the nonlinear saturation effects of heterogeneous atmospheric chemical reactions; on the other hand, upper-air remote sensing data is highly susceptible to local observation noise, leading to severe distortion in the quantification of coupling strength by a single algorithm. Existing evaluation systems lack a two-stage extraction mechanism capable of decoupling and recombining "optimal temporal locking" and "nonlinear intensity quantification," and also struggle to intuitively characterize the strength and direction of physical effects on continuous vertical profiles. Summary of the Invention

[0005] The purpose of this application is to provide a method, device and medium for quantitative assessment of spatial ozone distribution during dust storms. It realizes automated spatiotemporal resolution alignment of multi-source heterogeneous vertical observation data, effectively removes background noise from the diurnal variation of ozone itself, improves the signal-to-noise ratio of dust storm disturbance signals, and significantly enhances the statistical robustness and physical accuracy of the quantitative assessment results.

[0006] To achieve the above objectives, this application provides the following solution.

[0007] Firstly, this application provides a method for quantitatively assessing the spatial ozone distribution during dust storms, including: Obtain the two-dimensional observation matrix of dust characterization and ozone characterization during the dust event in the target area; By unifying the time base and resampling the spatiotemporal grid of the two-dimensional observation matrix for dust characterization and the two-dimensional observation matrix for ozone characterization, a dust observation matrix and an ozone observation matrix with completely consistent spatiotemporal resolution are obtained. The ozone anomaly matrix is ​​determined based on the ozone observation matrix and the average ozone concentration values ​​at corresponding times of typical clean days. The dust observation matrix and ozone anomaly matrix are decomposed into multiple sequence pairs according to the height dimension. For each sequence pair, a sliding correlation analysis is performed along the time axis within a preset time shift range. The Pearson correlation coefficient at each sliding position is calculated, and the time shift corresponding to the maximum absolute value of the Pearson correlation coefficient is determined as the optimal time shift for each height layer. Under the optimal time shift, the time axis-aligned sequence pairs are extracted, and the Spearman rank correlation coefficient between the two is calculated. The quantitative assessment results of dust-ozone coupling are generated based on the optimal temporal offset and Spearman rank correlation coefficient of all height layers. The quantitative assessment results of dust-ozone coupling include a continuous vertical profile with height as the ordinate and time offset as the abscissa, and a composite layer obtained by scalar color mapping of each node or segment on the continuous vertical profile based on the Spearman rank correlation coefficient.

[0008] Secondly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to implement the steps of the above-described method for quantitatively assessing spatial ozone distribution during sandstorm passage.

[0009] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method for quantitatively assessing spatial ozone distribution during sandstorm passage.

[0010] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a method, device, and medium for quantitative assessment of spatial ozone distribution during dust storms. By unifying the time reference and resampling the spatiotemporal grid of the two-dimensional observation matrix representing dust and the two-dimensional observation matrix representing ozone, a dust observation matrix and an ozone observation matrix with completely consistent spatiotemporal resolution are obtained. This solves the problem of spatiotemporal resolution mismatch in multi-source heterogeneous matrices in traditional methods, achieving automated spatiotemporal resolution alignment of multi-source heterogeneous vertical observation data. Based on the ozone observation matrix and the average ozone concentration values ​​at corresponding times on typical clean days, an ozone anomaly matrix is ​​determined, effectively removing the diurnal variation background noise of ozone itself and improving the signal-to-noise ratio of dust disturbance signals. The dust observation matrix and the ozone anomaly matrix are decomposed into multiple sets of one-dimensional time series pairs according to the height dimension. For each set of series pairs, a sliding correlation analysis is performed along the time axis within a preset time shift range, calculating the Pearson correlation coefficient at each sliding position, and determining the maximum absolute value of the Pearson correlation coefficient. The time shift corresponding to each time level is used as the optimal time offset for each altitude layer. Under the optimal time offset, time axis-aligned sequence pairs are extracted, and the Spearman rank correlation coefficient between the two is calculated. A two-order cross-correlation extreme value extraction mechanism of "Pearson time series locking + Spearman intensity quantization" is proposed. This mechanism decouples the two tasks with different statistical requirements, namely "optimal time series offset locking" and "nonlinear coupling intensity quantization": the first step uses the high sensitivity of Pearson correlation coefficient for linear time series alignment to quickly lock the optimal offset; the second step, based on time series alignment, uses Spearman rank correlation coefficient (a nonparametric statistic) to overcome the quantization distortion problem of traditional Pearson correlation analysis under the nonlinear saturation effect of atmospheric heterogeneous chemical reactions and remote sensing observation noise interference. The algorithms of the two stages are specialized and interconnected. Compared with the general analysis of a single algorithm in correlation technology, it significantly improves the statistical robustness and physical accuracy of quantitative evaluation results. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly described below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort: Figure 1 This is an application environment diagram of a method for quantitatively assessing spatial ozone distribution during a sandstorm, as described in one embodiment of this application. Figure 2 A flowchart illustrating a method for quantitatively assessing spatial ozone distribution during a sandstorm, provided in an embodiment of this application; Figure 3 A schematic diagram illustrating the specific process of a method for quantitatively assessing spatial ozone distribution during sandstorm passage, provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0013] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0014] The method for quantitatively assessing spatial ozone distribution during dust storms provided in this application can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be set up independently, integrated into server 104, or placed in the cloud or on other servers. Terminal 102 can send the multi-source vertical raw data to be processed (including the two-dimensional observation matrix for dust and the two-dimensional observation matrix for ozone) to server 104. After receiving the multi-source vertical raw data, server 104 performs time base unification and spatiotemporal grid resampling on the two-dimensional observation matrices for dust and ozone. Based on the ozone observation matrix and typical clean days, it determines the ozone anomaly matrix. The dust observation matrix and the ozone anomaly matrix are decomposed into multiple sequence pairs according to height. For each sequence pair, a sliding correlation analysis is performed along the time axis to calculate the Pearson correlation coefficient and determine the optimal time offset for each height layer. Under the optimal time offset, the time axis-aligned sequence pairs are extracted, and the Spearman rank correlation coefficient is calculated. The optimal time offsets output from all height layers are stitched together along the height axis to generate a continuous vertical profile, and a composite layer is output based on the Spearman rank correlation coefficient. Server 104 can feed back the obtained composite layer for the multi-source vertical raw data to terminal 102. In addition, in some embodiments, the method for quantitatively assessing the spatial ozone distribution during sandstorms can also be implemented by either server 104 or terminal 102. For example, terminal 102 can directly perform quantitative assessment of spatial ozone distribution on the multi-source vertical raw data to be processed, or server 104 can obtain the multi-source vertical raw data to be processed from the data storage system and perform quantitative assessment of spatial ozone distribution on the multi-source vertical raw data to be processed.

[0015] The terminal 102 can be, but is not limited to, various desktop computers and laptops. The server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers, or it can be a cloud server.

[0016] In one exemplary embodiment, such as Figure 2 As shown, a method for quantitatively assessing spatial ozone distribution during sandstorms is provided. This method is executed by computer equipment, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps 201 to 205.

[0017] Step 201: Obtain the two-dimensional observation matrix of dust characterization and ozone characterization during the dust event in the target area. The two-dimensional observation matrices of dust characterization and ozone characterization are obtained by observation instruments observing the target area.

[0018] Step 202: Unify the time reference and resample the spatiotemporal grid of the two-dimensional observation matrix for dust characterization and the two-dimensional observation matrix for ozone characterization to obtain dust observation matrix and ozone observation matrix with completely consistent spatiotemporal resolution.

[0019] Step 203: Determine the ozone anomaly matrix based on the ozone observation matrix and the average ozone concentration values ​​at the corresponding time of a typical clean day.

[0020] Step 204: Decompose the dust observation matrix and ozone anomaly matrix into multiple sequence pairs according to the height dimension; for each sequence pair, perform sliding correlation analysis along the time axis within a preset time shift range, calculate the Pearson correlation coefficient at each sliding position, and determine the time shift corresponding to the maximum absolute value of the Pearson correlation coefficient as the optimal time offset for each height layer; under the optimal time offset, extract the sequence pairs that are aligned on the time axis and calculate the Spearman rank correlation coefficient between them.

[0021] Step 205: Generate the quantitative assessment results of dust-ozone coupling based on the optimal temporal offset and Spearman rank correlation coefficient of all height layers. The quantitative assessment results of dust-ozone coupling include a continuous vertical profile with height as the ordinate and time offset as the abscissa, and a composite layer obtained by scalar color mapping of each node or segment on the continuous vertical profile based on the Spearman rank correlation coefficient.

[0022] By implementing steps 201 to 205 above, and unifying the time base and resampling the spatiotemporal grid of the two-dimensional observation matrix representing dust and ozone, a dust observation matrix and an ozone observation matrix with completely consistent spatiotemporal resolution are obtained. This solves the problem of spatiotemporal resolution mismatch of multi-source heterogeneous matrices in traditional methods and realizes automated spatiotemporal resolution alignment of multi-source heterogeneous vertical observation data. Based on the ozone observation matrix and the average ozone concentration values ​​at corresponding times of typical clean days, an ozone anomaly matrix is ​​determined, effectively removing the diurnal variation background noise of ozone itself and improving the signal-to-noise ratio of dust disturbance signals. The dust observation matrix and the ozone anomaly matrix are decomposed into multiple sets of one-dimensional time series pairs according to the height dimension. For each set of series pairs, a sliding correlation analysis is performed along the time axis within a preset time shift range. The Pearson correlation coefficient at each sliding position is calculated, and the time shift corresponding to the maximum absolute value of the Pearson correlation coefficient is determined as the time shift for each set of series pairs. The optimal time series offset at a certain altitude is used to extract time-aligned sequence pairs and calculate their Spearman rank correlation coefficients. A two-stage cross-correlation extreme value extraction mechanism (two-stage decoupled extraction mechanism) is proposed, which combines Pearson time series locking with Spearman intensity quantization. This mechanism decouples the two tasks with different statistical requirements: the first stage uses the high sensitivity of the Pearson correlation coefficient to linear time series alignment to quickly lock the optimal offset; the second stage, based on time series alignment, uses the Spearman rank correlation coefficient (a nonparametric statistic) to overcome the quantization distortion problem of traditional Pearson correlation analysis under the nonlinear saturation effect of atmospheric heterogeneous chemical reactions and remote sensing observation noise interference. The two-stage algorithms are specialized and interconnected, which significantly improves the statistical robustness and physical accuracy of quantitative evaluation results compared to the general analysis of a single algorithm in correlation techniques.

[0023] like Figure 3 As shown, firstly, multi-source vertical raw data covering the target spatial range (i.e., the target area) are acquired. Then, two-dimensional observation matrices for dust characterization and ozone characterization covering the target spatial range (e.g., 0-3000m height) are obtained from the observation instruments. The two-dimensional observation matrices are structured first-order physical quantity matrices arranged in a height-time grid, processed by the instrument terminal inversion algorithm. Specifically, the dust characterization two-dimensional observation matrix consists of gridded data of aerosol extinction coefficients or backscattering coefficients; the ozone characterization two-dimensional observation matrix consists of gridded data of ozone concentration (or mixing ratio) profiles. The dust characterization two-dimensional observation matrix can be gridded data of extinction coefficients retrieved from a lidar, and the ozone characterization two-dimensional observation matrix can be ozone lidar or radiosonde profile data. The original dimensions of the two matrices are: Dust Characterization Two-Dimensional Observation Matrix X raw ( i x , jx ),in i x This serves as the index for the original time sampling points in the two-dimensional observation matrix representing dust. j x The original height sampling point index for the two-dimensional observation matrix characterizing dust; the two-dimensional observation matrix O for ozone characterization. 3raw ( i o , j o ),in i o This is the index of the original time sampling points for the two-dimensional observation matrix characterizing ozone. j o This is the index of the original height sampling points for the two-dimensional observation matrix characterizing ozone.

[0024] Step 202 is executed by a computer or data processing server. Its purpose is to eliminate the differences in sampling frequency and vertical spatial resolution between different observation instruments, as well as the boundary truncation effect caused by the subsequent sliding algorithm, and to achieve standardized grid alignment of multi-source raw observation data.

[0025] In step 202 above, the two-dimensional observation matrix for dust characterization and the two-dimensional observation matrix for ozone characterization are unified in terms of time reference and resampled in terms of spatiotemporal grid to obtain dust observation matrix and ozone observation matrix with completely consistent spatiotemporal resolution. Specifically, this includes the following steps 2021 to 2022.

[0026] Step 2021, Time Standardization: The timestamps of the two-dimensional observation matrices for dust and ozone are standardized to local standard time, and the time resolution is standardized to a preset reference resolution. Specifically, the timestamps of all observation data (i.e., the two-dimensional observation matrices for dust and ozone) are forcibly converted to local standard time (LST), and the time resolution is standardized to a preset reference resolution (e.g., 5 minutes, 10 minutes, or 15 minutes).

[0027] Set core event time window [T] start , T end and maximum permissible translation step size For the dust storm observation matrix, data is strictly extracted within the core event time window; for the ozone observation matrix, data is extracted within an extended time window including the preceding and following buffer periods; the core event time window (starting time T) start With the termination time T end The criteria for determining the value of dust are: within the monitoring height range (e.g., 0–3000 m), a number of consecutive (N) characteristic physical quantities (e.g., extinction coefficient) in the two-dimensional observation matrix characterizing dust are within a certain range. tThe start and end times of each time grid point exceeding the preset dust identification threshold are used. For the dust observation matrix X, data within the time window T is strictly truncated; for the ozone observation matrix O3, an asymmetric time window including the preceding and following buffer periods is truncated. The data is used to ensure the integrity of the boundary sequence during subsequent sliding calculations and to avoid edge truncation effects.

[0028] Step 2022, Spatial grid resampling: The two-dimensional observation matrix for dust characterization and the two-dimensional observation matrix for ozone characterization are resampled on the time axis and the height axis according to a preset unified grid to generate dust observation matrix and ozone observation matrix with completely consistent spatiotemporal resolution.

[0029] Spatiotemporal grid resampling uses a bilinear interpolation algorithm, and the calculation formula is as follows: ; in, To unify the gridding, the dust monitoring matrix or ozone monitoring matrix at the target time grid points and height grid layer The target value for resampling at the location; and The time immediately preceding the target calculation time on the original timeline. And satisfy ≤ ≤ Two adjacent original time sampling points; and On the original height axis, the calculated height h immediately adjacent to the target and satisfying... ≤ ≤ Two adjacent original height sampling points; The actual observed values ​​of the original two-dimensional observation matrix at the corresponding original sampling points (corresponding to X in the text). raw Or O 3raw ); For the two-dimensional observation matrix characterizing dust or the two-dimensional observation matrix characterizing ozone at the original time sampling points and original height sampling point The actual observed values ​​at the location; For the two-dimensional observation matrix characterizing dust or the two-dimensional observation matrix characterizing ozone at the original time sampling points and original height sampling point The actual observed values ​​at the location; For the two-dimensional observation matrix characterizing dust or the two-dimensional observation matrix characterizing ozone at the original time sampling points and original height sampling point The actual observed values ​​at the location; For the two-dimensional observation matrix characterizing dust or the two-dimensional observation matrix characterizing ozone at the original time sampling points and original height sampling point The actual observed values ​​at that location.

[0030] The original matrices of both dimensions are resampled along the time and height axes according to a pre-defined uniform grid (e.g., the height axis is layered at 75m intervals), generating dust observation matrices X(h, t) and ozone observation matrices O3(h, t) with completely consistent spatiotemporal resolution. Here, h represents the uniform height grid layer index (h = 1, 2, …, H, where H is the total number of height layers), and t represents the uniform time grid point index (t = 1, 2, …, N, where N is the total number of sampling points within the time window). After this step, the row and column dimensions of the two matrices are completely consistent, supporting pixel-level point-by-point operations.

[0031] Under natural conditions, ozone concentration is influenced by solar radiation, exhibiting a strong periodic pattern of "higher during the day and lower at night." When dust storms pass through, their actual impact on ozone is superimposed on this natural "background fluctuation." If raw observations are used directly for analysis, the algorithm cannot distinguish whether the ozone fluctuations originate from conventional photochemical reactions or the physicochemical effects of dust storms, leading to statistical distortion. Therefore, step 203 uses data from a "typical clean day" without dust storm interference to construct a static background baseline, providing a corresponding background reference baseline for each altitude layer. By subtracting this baseline from the actual observation data, the system effectively filters out conventional diurnal background noise and extracts the ozone anomaly that mainly characterizes the impact of dust storm disturbances. This data purification process is a core prerequisite for ensuring the accuracy and physical significance of subsequent sliding analysis results. Step 203 includes steps 2031 to 2032.

[0032] Step 2031: Construct a two-dimensional background baseline: Extract several typical clean days on the time axis that are strictly independent of the event window and the preceding buffer period. The quantitative screening criteria for typical clean days are: the daily average PM10 concentration near the ground is lower than the daily average PM10 concentration limit in the Class I Ambient Air Quality Standard (50 μg / m³). 3 Furthermore, the aerosol optical depth (AOD) or the full profile extinction coefficient does not show a significant increase during the corresponding time period (no significant increase means that the difference between the aerosol optical depth or the full profile extinction coefficient at two moments within the corresponding time period is within a set range), ensuring no interference from external dust or localized heavy pollution air masses. For each altitude layer h and each corresponding hourly time or time grid point t, the average ozone concentration value at the corresponding moment of the clean day is calculated, thereby generating a highly dependent static background spatiotemporal distribution map matrix. .

[0033] ; in For the unified height grid layer index, For the unified time grid point index, This represents the total sample size for a typical clean day. For the summation index variable of a typical clean day, Then it means the first In a typical clean day, at a high level and time The actual observed ozone concentration. The rows of this static background spatiotemporal distribution map matrix correspond to the altitude layer, and the columns correspond to the time grid points. Its physical meaning is: the climatological expectation of ozone concentration in the altitude-time two-dimensional space under natural conditions without dust events.

[0034] Step 2032, Extract the pure anomaly matrix: Subtract the corresponding background matrix from the ozone observation matrix during the dust storm event to obtain the ozone anomaly matrix, i.e.: ; in, Indicates at altitude ,time An abnormal increase (positive) or decrease (negative) in ozone concentration caused by dust disturbance, expressed in the original units of ozone concentration (e.g., μg / m³). 3 The ozone anomaly matrix is ​​constructed from the anomalous increases or decreases in ozone concentration at all altitudes and at all times. The core function of this step is to remove the "background noise" of the photochemical diurnal variation of ozone itself, so that subsequent correlation calculations can focus on the pure disturbance signal introduced by dust.

[0035] The purpose of step 204 is to continuously extract the optimal temporal offset between the evolution characteristics of dust and ozone at different altitude layers and accurately quantify their nonlinear coupling strength. This step is the core innovation of this application, which distinguishes it from existing single sliding correlation analysis. It adopts a two-stage decoupling extraction mechanism of "Pearson temporal locking + Spearman intensity quantization".

[0036] 1) Height slicing: The algorithm combines the dust observation matrix X(h, t) and the ozone anomaly matrix. The system is divided layer by layer according to the height grid h, and transformed into H sets of one-dimensional time series pairs. That is, for each height layer h, the dust time series vector x is extracted. h = [x h (1), x h (2), …, x h (N1)] and ozone anomaly time series vector y h = [y h (1), yh (2),…, y h (M1)]. Here, N1 is the length of the dust time series vector, and M1 is the length of the ozone anomaly time series vector, with M1 > N1 (because the ozone sequence includes preceding and following buffer periods). The purpose of converting the two-dimensional matrix into a one-dimensional time series pair for comparison is to achieve dimensionality reduction and decoupling in spatial dimensions, independently extracting the temporal characteristics of dust and ozone evolution at a single altitude layer. The one-dimensional time series pair generated in this step is the basic data structure for subsequent two-order sliding analysis, directly used for translational traversal to lock the optimal temporal offset for each altitude layer, and further quantifying its nonlinear coupling strength based on this.

[0037] 2) First-order – Pearson time-series locking: For each altitude layer h, the shorter dust time series vector x h In the longer time series vector y of ozone anomalies h Bidirectional translation is performed within the entire window, and due to the buffer period setting, the amount of overlapping data during the bidirectional translation process remains constant.

[0038] A sliding correlation analysis is performed along the time axis within a preset time shift range to calculate the Pearson correlation coefficient at each sliding position. Specifically, this includes: for each altitude layer, extracting the dust time series vector and the ozone anomaly time series vector respectively; for each altitude layer, bidirectionally translating the dust time series vector within the complete window of the ozone anomaly time series vector, traversing all possible translation step sizes, and in each translation overlap, extracting the overlapping window of the two sequences that are aligned in time, and calculating the Pearson correlation coefficient within each overlapping window; the Pearson correlation coefficient within the overlapping window is the Pearson correlation coefficient at the sliding position.

[0039] Let the translation step size be... ( = 0, ±1, ±2, …, ± (corresponding index steps), in each translation overlap, a window of overlapping portion of the two sequences aligned in time is extracted, denoted as . and Calculate the Pearson correlation coefficient within the overlapping window, denoted as . The formula for calculating the Pearson correlation coefficient is as follows: ; in, For height layer And the translation step size is At that time, the Pearson correlation coefficient between the dust storm time series vector and the ozone anomaly time series vector; This represents the total number of valid sampling points within the overlapping window. and These are the truncated and aligned overlapping sequences of sand and dust. Ozone overlapping sequence In the The specific value at the nth sampling point, i.e. the nth Dust extinction coefficient and ozone anomaly at each sampling point; and These are the overlapping sequences of dust particles during this translation. Ozone overlapping sequence The arithmetic mean within the currently overlapping window.

[0040] Iterate through all possible translation step sizes k, and take the extreme value of the Pearson correlation coefficient, that is, extract the Pearson correlation coefficient with the largest absolute value |r max | and directly lock the extreme value|r max |Corresponding optimal timing offset : ; in The unified time grid step size, To determine the optimal translation step size, under this mathematical definition, if >0, physically meaning that changes in dust characteristics precede anomaly responses to ozone (i.e., ozone responses exhibit a lag); if A value <0 indicates that the ozone anomaly response precedes changes in dust characteristics. The purpose of using the Pearson correlation coefficient here is that it has high sensitivity to linear time series alignment, making it suitable for quickly identifying the optimal time series offset.

[0041] 3) Second-order Spearman intensity quantization: After determining the optimal time offset τ(h), the window sequence of the overlapping part of the dust time series vector and the ozone anomaly time series vector, which are strictly aligned with the time axis, is extracted using the above-mentioned optimal translation step size kopt.

[0042] Under the optimal time offset, the time axis-aligned sequence pairs are extracted, and the Spearman rank correlation coefficient between the two is calculated. Specifically, the overlapping window sequence of the dust time series vector and the ozone anomaly time series vector with the optimal translation step size is extracted, and the Spearman rank correlation coefficient between the two is calculated. For each altitude layer, its optimal time offset and the signed Spearman rank correlation coefficient are recorded to form a three-dimensional mapping relationship of altitude layer-offset-correlation coefficient.

[0043] For the aligned sequence pairs described above, calculate their Spearman's Rank Correlation Coefficient, denoted as [Formula omitted for brevity]. The formula for calculating the Spearman rank correlation coefficient is as follows: ; in, For height layer Furthermore, after the time axis has been strictly aligned using the optimal translation step size, the Spearman rank correlation coefficient between the dust time series vector and the ozone anomaly time series vector is obtained. For the first At each sampling point, the difference between the extinction coefficient of dust and the ozone anomaly in the current sequence window after sorting by numerical value. ; This represents the total number of valid sampling points within the overlapping window. When identical observation values ​​exist within the current sequence window, the identical observation values ​​are sorted by average rank.

[0044] The Spearman rank correlation coefficient ranges from -1 to 1. A positive value indicates a positive correlation between increased dust extinction coefficient and increased ozone anomaly (i.e., dust promotes ozone formation), while a negative value indicates a negative correlation between increased dust extinction coefficient and decreased ozone anomaly (i.e., dust inhibits ozone formation). The purpose of using the Spearman rank correlation coefficient here is that it is a nonparametric statistic, independent of linear assumptions, effectively avoids the nonlinear saturation effect of heterogeneous atmospheric chemical reactions, and is robust to local observation noise.

[0045] 4) Data Recording: For each height layer h, record its optimal temporal offset τ(h) and the signed Spearman rank correlation coefficient. (within the range [-1, 1]), forming a three-dimensional mapping relationship of height layer – offset – correlation coefficient, which serves as a scalar data source for subsequent visualization.

[0046] The purpose of step 205 is to transform the matrix operation results into an intuitive visualization of dust-ozone evolution, enabling researchers to understand at a glance the continuous evolution characteristics of the dust-ozone coupling effect in the vertical direction.

[0047] 1) Profile generation: Optimal temporal offset of all height layers output Re-stitch along the height Y-axis to generate a continuous vertical profile with height as the ordinate and time offset as the abscissa (i.e., Figure 3(Continuous vertical phase difference profile in the image). The original discrete points can be further processed into smooth curves by smoothing splines or local weighted regression (LOESS) to eliminate high-frequency fluctuations introduced by layer-by-layer calculations and to more clearly present the systematic change trend in the vertical direction.

[0048] 2) Composite Layer Output: In the final layer visualization, the system uses the height layer (Y-axis, unit: m) as the vertical axis and the optimal temporal offset. A two-dimensional coordinate system is constructed with the x-axis (unit: min or h) as the horizontal axis. Simultaneously, the signed Spearman rank correlation coefficient retained in step 204 is used... This involves performing a standard continuous scalar color mapping on each node or segment of a continuous vertical profile. For example, it can generate warm and cool gradient color bands: warm colors (such as red) represent positive correlations (…). >0, dust storms promote ozone formation), cool colors (such as blue) indicate a negative correlation ( <0 indicates that dust inhibits ozone formation. The saturation or brightness of the color represents the absolute value of the correlation coefficient (i.e., the coupling strength). The offset distance (X-axis position) of the points in the figure characterizes the temporal misalignment of the dust-ozone coupling effect at different altitudes. A positive offset indicates that dust changes precede ozone responses, while a negative offset indicates that ozone anomalies precede dust changes. The mapped color visually represents the strength and direction of the physical effect.

[0049] This application also provides an application scenario in which the aforementioned quantitative assessment method for spatial ozone distribution during dust storms is applied. Specifically, the quantitative assessment method for spatial ozone distribution during dust storms provided in this embodiment can be applied to the diagnosis of atmospheric complex pollution events. The atmospheric complex pollution event diagnosis scenario includes a data acquisition stage, a spatial ozone distribution quantitative assessment link, and a content distribution stage. Multi-source vertical raw data enters the spatial ozone distribution quantitative assessment link from the data acquisition stage, obtains the corresponding composite layer through human-machine collaboration, and then enters the downstream atmospheric complex pollution event diagnosis stage. The quantitative assessment method for spatial ozone distribution during dust storms provided in this embodiment belongs to the spatial ozone distribution quantitative assessment link. Specifically, in the process of quantitatively assessing the spatial ozone distribution of multi-source vertical raw data, the two-dimensional observation matrix for dust and the two-dimensional observation matrix for ozone can be unified in terms of time reference and resampled in terms of spatiotemporal grid. Based on the ozone observation matrix and typical clean days, the ozone anomaly matrix is ​​determined. The dust observation matrix and the ozone anomaly matrix are decomposed into multiple sets of sequence pairs according to height. For each set of sequence pairs, sliding correlation analysis is performed along the time axis to calculate the Pearson correlation coefficient and determine the optimal time offset for each height layer. Under the optimal time offset, the time axis-aligned sequence pairs are extracted, and the Spearman rank correlation coefficient is calculated. The optimal time offsets output from all height layers are stitched together along the height axis to generate a continuous vertical profile, and a composite layer is output by combining the Spearman rank correlation coefficient.

[0050] This application has the following beneficial effects: 1) Automated spatiotemporal resolution alignment of multi-source heterogeneous vertical observation data is achieved. This application systematically solves the inconsistency in sampling frequency and vertical resolution between different observation platforms such as dust lidar and ozone lidar / sonsonde through time base unification and spatiotemporal grid resampling (interpolation algorithm unification gridding) in step one. Compared with the manual alignment or simple coarse-grained averaging methods in related technologies, this application achieves fully automated processing while ensuring data accuracy, significantly improving data processing efficiency and eliminating errors caused by the sliding calculation boundary truncation effect.

[0051] 2) This method effectively removes the diurnal variation background noise of ozone itself, improving the signal-to-noise ratio of dust disturbance signals. By constructing and subtracting the highly dependent two-dimensional static background baseline in step two, this application decomposes the temporal variation of ozone concentration into two physical components: "climatological diurnal variation background" and "dust event disturbance anomaly." Compared to the approach of directly performing correlation analysis on the absolute values ​​of observations using other techniques, the ΔO3(h, t) pure anomaly matrix extracted by this method more purely reflects the true increase or decrease effect of dust passage on ozone concentration, making the results of downstream correlation analysis more clearly and reliably interpreted physically.

[0052] 3) An innovative two-stage cross-correlation extreme value extraction mechanism, combining Pearson temporal locking and Spearman intensity quantization, is proposed. This mechanism decouples two tasks with different statistical requirements: "optimal temporal offset locking" and "nonlinear coupling intensity quantization." The first stage utilizes the high sensitivity of the Pearson correlation coefficient for linear temporal alignment to quickly lock the optimal offset. The second stage, based on temporal alignment, uses the Spearman rank correlation coefficient (a nonparametric statistic) to overcome the quantization distortion problems of traditional Pearson correlation analysis under the nonlinear saturation effect of atmospheric heterogeneous chemical reactions and remote sensing noise interference. The two-stage algorithms each have their own specialization and are interconnected, significantly improving the statistical robustness and physical accuracy of the quantitative evaluation results compared to the general analysis of a single algorithm in correlation techniques.

[0053] 4) It provides continuous vertical profile visualization output, enabling an intuitive representation of the dust-ozone coupling effect in the vertical dimension. Through profile reconstruction and composite layer mapping in step four, this application integrates information from three dimensions—height layer (Y-axis), optimal temporal offset (X-axis), and coupling strength (color mapping)—into a single two-dimensional image. This allows researchers to intuitively identify the vertical evolution characteristics of the dust-ozone coupling effect at different height layers in terms of both "temporal misalignment" and "physical strength." The positive / negative distinction between cool and warm tones makes the promoting or inhibiting effects of dust readily apparent, significantly improving the scientific readability and decision support efficiency of the data analysis results.

[0054] 5) The method is highly universal and easy to promote and deploy in engineering. The technical solution proposed in this application does not rely on specific types of observation instruments or data characteristics of specific geographical areas, and has good versatility. As long as a multi-source vertical observation data matrix that meets the format requirements is provided, standardized evaluation results can be output through the above-mentioned automated process. This solution is applicable to various application scenarios such as regional dust-ozone coupling mechanism research, diagnosis of atmospheric compound pollution events, and calibration of air quality model parameterization schemes.

[0055] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 4As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs in the non-volatile storage media to run. The database stores a two-dimensional observation matrix for dust characterization, a two-dimensional observation matrix for ozone characterization, an ozone anomaly matrix, optimal time-series offsets, Spearman rank correlation coefficients, and composite layers. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for quantitatively assessing spatial ozone distribution during dust storms.

[0056] Figure 4 The structures shown are merely block diagrams of some structures related to the present application and do not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0057] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0058] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with relevant regulations and be authorized by the owner of the corresponding device.

[0059] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0060] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0061] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0062] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for quantitatively assessing spatial ozone distribution during sandstorms, characterized in that, The method includes: Obtain the two-dimensional observation matrix of dust characterization and ozone characterization during the dust event in the target area; By unifying the time base and resampling the spatiotemporal grid of the two-dimensional observation matrix for dust characterization and the two-dimensional observation matrix for ozone characterization, a dust observation matrix and an ozone observation matrix with completely consistent spatiotemporal resolution are obtained. The ozone anomaly matrix is ​​determined based on the ozone observation matrix and the average ozone concentration values ​​at corresponding times of typical clean days. The dust observation matrix and ozone anomaly matrix are decomposed into multiple sequence pairs according to the height dimension. For each sequence pair, a sliding correlation analysis is performed along the time axis within a preset time shift range. The Pearson correlation coefficient at each sliding position is calculated, and the time shift corresponding to the maximum absolute value of the Pearson correlation coefficient is determined as the optimal time shift for each height layer. Under the optimal time shift, the time axis-aligned sequence pairs are extracted, and the Spearman rank correlation coefficient between the two is calculated. The quantitative assessment results of dust-ozone coupling are generated based on the optimal temporal offset and Spearman rank correlation coefficient of all height layers. The quantitative assessment results of dust-ozone coupling include a continuous vertical profile with height as the ordinate and time offset as the abscissa, and a composite layer obtained by scalar color mapping of each node or segment on the continuous vertical profile based on the Spearman rank correlation coefficient.

2. The method for quantitatively assessing spatial ozone distribution during sandstorms according to claim 1, characterized in that, By unifying the time base and resampling the spatiotemporal grids of the two-dimensional observation matrices representing dust and ozone, dust and ozone observation matrices with completely consistent spatiotemporal resolutions are obtained, specifically including: The timestamps of the two-dimensional observation matrix for dust characterization and the two-dimensional observation matrix for ozone characterization are unified to the local standard time, and the time resolution is unified to the preset reference resolution. Set the core event time window and the maximum allowable translation step size; for the dust observation matrix, strictly extract the data within the core event time window; for the ozone observation matrix, extract the data within the extended time window including the buffer periods before and after; the determination criteria for the core event time window are: within the monitoring height range, the start and end times of the characteristic physical quantity in the two-dimensional dust characterization observation matrix exceeding the preset dust identification threshold for several consecutive time grid points. The two-dimensional observation matrix for dust and ozone is resampled along the time and height axes according to a preset uniform grid to generate dust and ozone observation matrices with completely consistent spatiotemporal resolution.

3. The method for quantitatively assessing spatial ozone distribution during sandstorm passage according to claim 1, characterized in that, Spatiotemporal grid resampling uses a bilinear interpolation algorithm, and the calculation formula is as follows: ； in, To unify the gridding, the dust monitoring matrix or ozone monitoring matrix at the target time grid points and height grid layer The target value for resampling at the location; and The time immediately preceding the target calculation time on the original timeline. And satisfy ≤ ≤ Two adjacent original time sampling points; and On the original height axis, the calculated height h immediately adjacent to the target and satisfying... ≤ ≤ Two adjacent original height sampling points; For the two-dimensional observation matrix characterizing dust or the two-dimensional observation matrix characterizing ozone at the original time sampling points and original height sampling point The actual observed values ​​at the location; For the two-dimensional observation matrix characterizing dust or the two-dimensional observation matrix characterizing ozone at the original time sampling points and original height sampling point The actual observed values ​​at the location; For the two-dimensional observation matrix characterizing dust or the two-dimensional observation matrix characterizing ozone at the original time sampling points and original height sampling point The actual observed values ​​at the location; For the two-dimensional observation matrix characterizing dust or the two-dimensional observation matrix characterizing ozone at the original time sampling points and original height sampling point The actual observed values ​​at that location.

4. The method for quantitatively assessing spatial ozone distribution during sandstorm passage according to claim 1, characterized in that, The quantitative screening criteria for a typical clean day are: the daily average concentration of PM10 near the ground is lower than the daily average concentration limit of PM10 in the first level of ambient air quality standard, and the aerosol optical depth or total profile extinction coefficient does not significantly increase in the corresponding period, ensuring that there is no external dust or local heavy pollution air mass interference; the first level of ambient air quality standard is 50 μg / m 3 .

5. The method for quantitatively assessing spatial ozone distribution during sandstorm passage according to claim 1, characterized in that, A sliding correlation analysis is performed along the time axis within a preset time shift range, and the Pearson correlation coefficient at each sliding position is calculated, specifically including: For each altitude level, extract the time series vectors of dust and ozone anomalies respectively; For each altitude layer, the dust time series vector is bidirectionally shifted within the complete window of the ozone anomaly time series vector, traversing all possible shift steps. In each shift overlap, the overlapping window of the two sequences that are aligned in time is extracted, and the Pearson correlation coefficient within each overlapping window is calculated. The Pearson correlation coefficient within the overlapping window is the Pearson correlation coefficient at the sliding position.

6. The method for quantitatively assessing spatial ozone distribution during sandstorms according to claim 1, characterized in that, At the optimal time offset, extract time-aligned sequence pairs and calculate their Spearman rank correlation coefficient, specifically including: Using the optimal translation step size, the overlapping window sequence of the dust time series vector and the ozone anomaly time series vector with strictly aligned time axes is extracted, and the Spearman rank correlation coefficient between the two is calculated. For each height layer, its optimal temporal offset and signed Spearman rank correlation coefficient are recorded to form a three-dimensional mapping relationship between height layer, offset, and correlation coefficient.

7. The method for quantitatively assessing spatial ozone distribution during sandstorm passage according to claim 5, characterized in that, The formula for calculating the Pearson correlation coefficient is as follows: ； in, For height layer And the translation step size is At that time, the Pearson correlation coefficient between the dust storm time series vector and the ozone anomaly time series vector; This represents the total number of valid sampling points within the overlapping window. and These are the truncated and aligned overlapping sequences of sand and dust. Ozone overlapping sequence In the Dust extinction coefficient and ozone anomaly at each sampling point; and These are the overlapping sequences of dust particles during this translation. Ozone overlapping sequence The arithmetic mean within the currently overlapping window.

8. The method for quantitatively assessing spatial ozone distribution during sandstorm passage according to claim 6, characterized in that, The formula for calculating the Spearman rank correlation coefficient is as follows: ； in, For height layer Furthermore, after the time axis has been strictly aligned using the optimal translation step size, the Spearman rank correlation coefficient between the dust time series vector and the ozone anomaly time series vector is obtained. For the first At each sampling point, the difference between the extinction coefficient of dust and the ozone anomaly in the current sequence window after sorting by numerical value. ; This represents the total number of valid sampling points within the overlapping window. When identical observation values ​​exist within the current sequence window, the identical observation values ​​are sorted by average rank.

9. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that the processor executes the computer program to implement the method for quantitative assessment of spatial ozone distribution during dust storms as described in any one of claims 1-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the method for quantitative assessment of spatial ozone distribution during sandstorm passage as described in any one of claims 1-8.

Images