Method for reconstructing regional automatic station temperature sequence based on scale separation and meteorological geographical zoning

Abstract

The application discloses a regional automatic station air temperature sequence reconstruction method based on scale separation and meteorological geographical zoning, and comprises the following steps: performing scale separation on regional automatic station observation data to obtain a regional automatic station climatic sequence and a regional automatic station anomaly sequence; performing scale separation on observation data of a national standard meteorological station to obtain a national station climatic sequence; combining the two climatic sequences to obtain a fitting climatic sequence; performing scale separation on remote sensing data to obtain a remote sensing data anomaly sequence; combining the two anomaly sequences to obtain a fitting anomaly sequence; and performing sequence synthesis on the fitting climatic sequence and the fitting anomaly sequence to obtain a regional automatic station reconstruction sequence.

Description

Technical Field

[0001] This invention relates to the technical field of regional automatic weather station data processing, and specifically to a method for reconstructing regional automatic weather station temperature sequences based on scale separation and meteorological geographic zoning. Background Technology

[0002] The identification of extreme weather and climate events relies primarily on direct observation data from ground-based meteorological stations, which can be categorized into two types: national standard meteorological stations and regional automatic meteorological observation stations. National standard meteorological stations require real-time human monitoring and have strict requirements regarding their construction areas. Regional automatic stations, due to their automated observation capabilities, can conduct continuous observations even in sparsely populated areas, and their high station density compensates for the sparse distribution of traditional national standard meteorological stations. However, because regional observation stations are unmanned, their maintenance frequency is lower than that of national standard meteorological stations, leading to data gaps for certain periods when instruments malfunction or are damaged. Furthermore, some regional observation stations are located near farmland, cities, or highways, where human activities frequently cause abrupt changes in observed values ​​during the observation period. This often results in data quality that does not meet the requirements for climate analysis and research, affecting the utilization rate of regional automatic station observation data. Reconstructing missing data sequences from regional automatic stations is an effective way to ensure refined meteorological analysis and research.

[0003] Because temperature changes relatively slowly over time and spatial gradients are small in most areas, it exhibits a certain degree of continuity in both space and time. This allows for the reconstruction of temperature data sequences with longer timescales and better continuity by reconstructing data from regional automatic weather stations. The primary challenge in reconstructing missing data sequences from automatic weather stations is the selection of reference stations. Reference stations are those with high-quality and continuous observational data selected for spatial interpolation, typically national standard meteorological stations. However, since the amount of data from automatic weather stations is far greater than that from national stations, fewer national stations are chosen as reference stations. Multiple regional automatic weather stations may select the same national standard meteorological station as a reference station, leading to high spatial similarity in the reconstructed data sequences. This limits the spatial identification of extreme weather and climate events in the reconstructed sequences.

[0004] Currently, the reconstruction of missing temperature data from regional stations primarily employs spatial interpolation and temporal fitting methods, such as inverse distance weighting, trend surface methods, spline function methods, and linear fitting methods. Spatial interpolation algorithms that do not consider temporal fitting typically use distance as the main interpolation parameter, interpolating the automatic weather station data spatially using distance weighting from surrounding reference stations. This method significantly reduces the spatial distribution of topographic variations and local climate characteristics; the true resolution of the interpolated data is not improved, merely a mathematical downscaling, which has significant limitations. Furthermore, spatial interpolation methods weaken the influence of topographic and underlying surface inhomogeneities on temperature, exhibiting a certain spatial smoothing effect and erasing local climate effects, thus mitigating the characteristics of extreme high and low temperatures to some extent. Fitting methods incorporating temporal variations, such as linear regression, are highly dependent on reference stations, leading to reconstructed data with highly consistent characteristics with the reference stations, failing to reflect actual temperature variations. The accuracy of different interpolation methods depends on both spatial and temporal scales, and different interpolation methods have varying advantages and disadvantages in different regions. This can lead to situations where, when performing interpolation over a large range, using the same method results in significant errors in certain areas.

[0005] A key objective of temperature data sequence reconstruction is to achieve refined identification of extreme temperature events such as high and low temperature variations, thereby extracting key spatiotemporal indicators of extreme event evolution. However, due to the limitations of the methods mentioned above, the reconstruction of automatic weather station temperature data sequences currently tends to weaken extreme climate signals. The resulting interpolated data shows high consistency with the temperature change characteristics of national stations, but spatially, it largely erases local temperature variations. This reduces the ability to identify refined extreme weather and climate events, negating the significance of using highly refined automatic weather station temperature data and negatively impacting research on risk assessment and early warning of extreme weather and climate events. Summary of the Invention

[0006] The present invention aims to at least partially solve one of the technical problems existing in the related art.

[0007] The purpose of this invention is to provide a method for reconstructing regional automatic weather station temperature sequences based on scale separation and meteorological geographic zoning. The original temperature sequences of regional automatic weather stations are separated into climatological sequences and anomaly sequences, representing decadal and short-term signals of temperature changes, respectively. For the climatological sequences, data from national stations within the same meteorological geographic zoning are fitted. For the anomaly sequences, high-resolution remote sensing surface temperature data are fitted. The two types of sequences are fitted separately and then synthesized to obtain the reconstructed regional automatic weather station temperature data sequences.

[0008] To achieve the above objectives, the present invention provides a method for reconstructing regional automatic weather station temperature sequences based on scale separation and meteorological geographic zoning, comprising the following steps:

[0009] S1. Select regional automatic weather stations within the study area whose observation data quality meets the requirements; and perform quality control on the observation data of the selected regional automatic weather stations.

[0010] S2. Scale separation is performed on the quality-controlled regional automatic station observation data to obtain the regional automatic station climatological sequence representing climatological scale temperature change, and the regional automatic station anomaly sequence obtained by subtracting the regional automatic station climatological sequence from the original temperature sequence.

[0011] S3. Based on climate-geographic zoning, select several national standard meteorological stations located in the same geographic zone as the regional automatic weather stations using the nearest distance method; perform scale separation on the observation data of the selected national standard meteorological stations to obtain the national station climatological sequence; then combine the regional automatic weather station climatological sequence and the national station climatological sequence to construct a linear multiple regression model and calculate the fitted climatological sequence.

[0012] S4. Based on the latitude and longitude information of the regional automatic weather station, find the grid point closest to the regional automatic weather station from the high-resolution remote sensing inversion surface temperature grid data. Using this grid point as the center, select grid data within a set range and average them to obtain remote sensing data. Perform scale separation on the remote sensing data to obtain the remote sensing data anomaly sequence. Combine the regional automatic weather station anomaly sequence and the surface data anomaly sequence to construct a linear regression model and calculate the fitted anomaly sequence.

[0013] S5. The fitted climatological sequence and the fitted anomaly sequence are combined to obtain the regional automatic station reconstruction sequence.

[0014] A further preferred technical solution of the present invention is that step S1 involves selecting regional automatic weather stations within the study area whose observation data quality meets the requirements; specifically:

[0015] Extract the construction information of regional automatic weather stations within the study area, determine the station locations and surrounding environment, and remove regional automatic weather stations used for traffic observation and large water body observation; and perform observation quantity statistics on the observation data of the remaining regional automatic weather stations, and remove regional automatic weather stations with data duration of less than 5 years and missing data exceeding 45% of the total data volume.

[0016] Preferably, step S1 involves quality control of the selected area's automatic weather station observation data; specifically, this includes:

[0017] (1) Temperature threshold check: Determine the upper and lower thresholds of annual temperature based on the historical temperature observations of the study area, and remove data from the automatic station observation data of the area that exceed the upper and lower thresholds of annual temperature;

[0018] (2) Statistically analyze the observation data of the national standard meteorological stations around the regional automatic stations, calculate the monthly temperature upper and lower thresholds, and remove the data in the regional automatic station observation data that exceed the monthly temperature upper and lower thresholds for the corresponding month;

[0019] (3) Based on the observation data of the national standard meteorological station, calculate the temperature change threshold for 24, 48 and 72 hours in the area where the regional automatic station is located, and remove the data in the regional automatic station observation data that exceed the temperature change threshold;

[0020] (4) Set the standard deviation of temperature change within a specified time period and remove data from the automatic station observation data of the area that exceed ±3 times the standard deviation within that time period.

[0021] Preferably, step S2 involves scale separation of the quality-controlled regional automatic weather station observation data to obtain regional automatic weather station climatological sequences representing climatological-scale temperature changes, and regional automatic weather station anomaly sequences obtained by subtracting the regional automatic weather station climatological sequences from the original temperature sequences; specifically:

[0022] S21, Assume there are a total of regional automatic stations. Year data, number The original temperature sequence for the year is ,calculate Annual daily climate average series:

[0023] ;

[0024] S22. Using the moving average method, a moving average is calculated on the daily climate average series, with a moving window of 31 days and a moving average of 1 day at a time. and By performing a cyclic sliding process, the regional automatic weather station climatological sequence is obtained, as follows:

[0025] ;

[0026] This is a regional automatic weather station climatological sequence.

[0027] S23. Subtract the regional automatic station climatological sequence from the original temperature sequence of the regional automatic stations to obtain the regional automatic station anomaly sequence. , is represented as:

[0028] ;

[0029] This is an automatic station distance sequence for the region.

[0030] Preferably, the scale separation of the observation data from the selected national standard meteorological station in step S3 and the scale separation of the remote sensing surface data in step S4 are performed using the same method as the scale separation of the quality-controlled regional automatic station observation data in step S2.

[0031] Preferably, in step S3, scale separation is performed on the observation data of the selected national standard meteorological stations to obtain the national station climatological sequence. Then, a linear multiple regression model was constructed using it and the climatological sequences from regional automatic weather stations to calculate the fitted climatological sequences. , is represented as:

[0032] ;

[0033] in, This is the national station climatological sequence of the k-th national standard meteorological station; is the regression coefficient.

[0034] Preferably, step S4 involves finding the grid point closest to the regional automatic weather station from the high-resolution remote sensing inversion surface temperature grid data based on the latitude and longitude information of the regional automatic weather station, and then averaging the grid data within a set range centered on that grid point to obtain remote sensing data; specifically:

[0035] Based on the latitude and longitude information of the regional automatic weather stations, the grid points closest to the regional automatic weather stations are found from the high-resolution remote sensing inversion grid data of land surface temperature. The remote sensing temperature data value extracted from this grid point is Select 25 grid points centered on this grid point. The daily remote sensing data is obtained by averaging the data from 25 grid points, and is represented as follows:

[0036] ;

[0037] The remote sensing data sequence is composed of daily remote sensing data. , where m is the total number of years in the remote sensing data sequence, and l is the total number of days in the m-th year.

[0038] Preferably, in step S4, scale separation is performed on the remote sensing data sequence to obtain the remote sensing data anomaly sequence. Then, a regression model is constructed by combining it with the regional automatic station anomaly sequence, and the fitted anomaly sequence is calculated. , is represented as:

[0039] ;

[0040] Where n0 and n1 are the fitting coefficients.

[0041] Preferably, step S5 synthesizes the fitted climatological sequence and the fitted anomaly sequence to obtain the regional automatic station reconstruction sequence, represented as:

[0042] ;

[0043] in, For the reconstruction sequence of regional automatic stations, To fit the anomaly sequence, To fit the climatological sequence; The number of days in a year, with a value from 1 to 366, where i represents the year.

[0044] Preferably, in the calculations of steps S2 to S5, each year is counted as 366 days as a leap year. If the number of observations on a certain date is less than 50% of the total number of observations for the year, the observation data for that date is recorded as missing, and all data for that date are not included in the calculation.

[0045] In another aspect, the present invention provides a non-transitory computer-readable storage medium having computer instructions stored thereon, which cause a computer to execute the above-described method for reconstructing regional automatic station temperature sequences based on scale separation and meteorological geographic zoning.

[0046] In another aspect, the present invention provides an electronic device, comprising: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus, and the processor calls logical instructions in the memory to execute the above-mentioned method for reconstructing regional automatic station temperature sequences based on scale separation and meteorological geographic zoning.

[0047] In another aspect, the present invention provides a computer program product comprising a computer program stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer performs the above-described method for reconstructing regional automatic station temperature sequences based on scale separation and meteorological geographic zoning.

[0048] Beneficial Effects: The regional automatic weather station temperature sequence reconstruction method based on scale separation and meteorological and geographical zoning of this invention separates the climatological temperature sequence from the short-term scale through scale separation. The climatological sequence is fitted using national standard meteorological station data, while the anomaly sequence is fitted using refined remote sensing inversion surface temperature data. Different representative data are selected for fitting at different scales, which not only ensures the large-scale decadal background characteristics but also preserves the influence of local meteorological and geographical environments. Combining the two for automatic weather station data sequence reconstruction yields results that better reflect the extreme characteristics of temperature changes, avoiding the smoothing effect of using only national stations for fitting extreme values. This method is more suitable for extracting the characteristics of extreme temperature events, making the fitted data sequence closer to the actual distribution of temperature at large and small scales.

[0049] This invention uses climate-geographic zoning to select reference stations instead of relying on the principle of proximity to select national meteorological stations. This ensures the consistency of the surrounding geographical environment between the two types of observation stations being fitted, and minimizes the difference in climate background between the reconstructed sequence and the original sequence caused by differences in the surrounding environment of the stations.

[0050] Because remote sensing data has high spatial resolution, each grid point has a certain spatial difference from the surrounding grid points. By averaging the remote sensing data over a small area to obtain surface data to replace the single grid point data for data reconstruction, the uniformity of the reconstructed data in terms of spatial range is ensured. Attached Figure Description

[0051] Figure 1 This is a flowchart of the method for reconstructing regional automatic weather station temperature sequences based on scale separation and meteorological geographic zoning according to the present invention;

[0052] Figure 2 This is the original temperature sequence from the regional automatic weather station in Example 1;

[0053] Figure 3 Reconstructing the temperature sequence from the regional automatic weather stations in Example 1;

[0054] Figure 4 The frequency distribution of the raw temperature data from the regional automatic weather stations in Example 1;

[0055] Figure 5 The frequency distribution of reconstructed temperature data from regional automatic weather stations in Example 1;

[0056] Figure 6 The spatial distribution map of daily maximum temperature in summer is plotted using regional automatic weather stations to reconstruct temperature sequences, as shown in Example 1. Detailed Implementation

[0057] 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, embodiments of this invention, and should not be construed as limiting the 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. In the description of this invention, it should be understood that the terminology used is for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0058] The following is combined with Figures 1-6 This invention describes a method for reconstructing regional automatic weather station temperature sequences based on scale separation and meteorological geographic zoning.

[0059] Example 1: This example provides a method for reconstructing regional automatic weather station temperature sequences based on scale separation and meteorological geographic zoning, such as... Figure 1 As shown, it includes the following steps:

[0060] S1. Select regional automatic stations within the study area whose observation data quality meets the requirements; and perform quality control on the observation data of the selected regional automatic stations.

[0061] In this embodiment, Jiangsu Province in eastern my country is selected as the study area. In each step of the calculation, each year is counted as 366 days as a leap year. If the number of observations on a certain date is less than 50% of the total number of years, the observation data on that date is recorded as missing, and all data for that date are not included in the calculation.

[0062] S11. Selection of Regional Automatic Stations. This step is to select suitable regional automatic stations for data sequence reconstruction. The station construction information is extracted, the station locations and surrounding environment are determined, and regional automatic stations specifically used for traffic observation or large water body observation are excluded, as their data is not suitable for identifying extreme high-temperature disasters. The number of observations from the selected automatic stations is statistically analyzed, and regional automatic stations with data durations less than 5 years or missing data exceeding 45% of the total data volume are excluded to avoid poor fitting results due to insufficient data.

[0063] S12. Perform quality control on the acquired regional automatic weather station observation data, removing outliers that may affect the reconstruction of the data sequence. Since most regional automatic weather station data is unattended and relies on automatic instrument observation, the stations may be affected by nearby human activities or natural phenomena (such as spontaneous combustion, anthropogenic heat sources, or artificial watering), leading to abnormal observations that do not match the actual atmospheric conditions. These outliers will affect the fitting model parameters and adversely impact the accuracy of the reconstructed data sequence. Outlier removal needs to be performed according to the following procedure:

[0064] (1) Temperature threshold check: my country has a vast territory and significant differences between the north and south. Based on historical temperature observations, temperature thresholds are set for different regions. The temperature range in eastern my country is [-50, 50]. Values ​​exceeding this range are considered outliers and are therefore removed. , If a certain observation data or Then the data will be removed. The threshold can be adjusted for different regions based on historical statistics.

[0065] (2) Monthly threshold check: Since temperature observation data have obvious annual cycle characteristics, monthly threshold removal can compress the extreme value range and make it easier to find outliers. Data from national stations surrounding the statistical area's automatic weather stations are used to calculate the maximum monthly temperature. and minimum value .

[0066] ;

[0067] ;

[0068] Where the subscript mon represents the month, i represents a day within that month (values ​​1-31), and j represents the year of observation (1-5 if there are 5 years of observations). This calculation method extracts the same month from all years and performs extreme value statistics, ensuring that the extreme value is the maximum or minimum value for that month over many years. Monthly temperature extreme values ​​are amplified by a factor of 1.5 to serve as a monthly threshold; values ​​exceeding this threshold are considered outliers and are removed. and The observation data from national standard meteorological stations surrounding the regional automatic weather stations were statistically analyzed, and the upper and lower thresholds for monthly temperatures were calculated. Data from the regional automatic weather stations that exceeded the upper and lower thresholds for the corresponding month were then removed.

[0069] (3) Temperature fluctuation threshold check: Since temperature changes have a certain continuity within a set time period, the range of temperature changes is checked based on this characteristic. Using data from national stations across the province, the maximum 24, 48, and 72-hour temperature fluctuation values ​​are calculated:

[0070] ;

[0071] ;

[0072] ;

[0073] Considering that the temperature fluctuation range of automatic weather stations is greater than that of national weather stations, the temperature change threshold of automatic weather stations is increased. , or At that time, remove Temperature value at any given time.

[0074] (4) Sliding standard deviation check: The sliding standard deviation method considers that within a set time range, the temperature threshold range does not exceed ±3 times the standard deviation within that time period. Within the sliding window range, the temperature value T should satisfy:

[0075] ;

[0076] ;

[0077] Data will be removed if this condition is not met. The sliding window length is set to 10-30 days, with each slide lasting one day. This represents the average value of the data within the sliding window range.

[0078] The frequency distribution of the raw temperature sequence and raw temperature data from the regional automatic weather stations in this embodiment is as follows: Figure 2 and Figure 4 As shown.

[0079] S2. Scale separation is performed on the quality-controlled regional automatic station observation data to obtain the regional automatic station climatological sequence representing climatological scale temperature changes, and the regional automatic station anomaly sequence obtained by subtracting the regional automatic station climatological sequence from the original temperature sequence.

[0080] S21. When calculating the climate background sequence, a data continuity check is required first, mainly to ensure the reliability of annual cycle data when separating time scales. If the monthly observation data is less than 50% or the annual data is less than 50%, then the monthly and annual data will not be included in the climate background sequence calculation.

[0081] S22. Climatological sequence calculation. The climatological sequence represents the temperature change sequence at the climatological scale, mainly showing the annual temperature cycle, while retaining temperature change information beyond monthly variations.

[0082] Assume there are a total of regional automatic stations Year data, number The original temperature sequence for the year is ,calculate Annual daily climate average series:

[0083] ;

[0084] S23. To filter out temporal variation signals below the monthly scale, a moving average method is used to perform a moving average on the daily climate average series. The moving window is 31 days, and each moving average is 1 day. and By performing a cyclic sliding process, the regional automatic weather station climatological sequence is obtained, as follows:

[0085] ;

[0086] This is a regional automatic weather station climatological sequence.

[0087] S24. Anomaly Series Calculation. The regional automatic weather station anomaly series is obtained by subtracting the regional automatic weather station climatological series from the original temperature series. , is represented as:

[0088] ;

[0089] This is an automatic station distance sequence for the region.

[0090] S3. Based on climate geographic zoning, select 1-5 national standard meteorological stations located in the same geographic zone as the regional automatic weather stations using the nearest distance method; perform scale separation on the observation data of the selected national standard meteorological stations to obtain the national station climatological series. Then, a linear multiple regression model was constructed using it and the climatological sequences from regional automatic weather stations to calculate the fitted climatological sequences. , is represented as:

[0091] ;

[0092] in, This is the national station climatological sequence of the k-th national standard meteorological station; The regression coefficients are estimated using the least squares method:

[0093] ;

[0094] Where X = F= ;

[0095] The existing regional automatic weather station climatological sequences The f-value is substituted into the coefficient formula to obtain the coefficients of each term. Then, the climatological sequence data from the national station is substituted into the equations that determine the coefficients. Calculate the fitted regional automatic weather station climatological sequence f. The fitted climatological sequence does not contain any missing data.

[0096] S4. Based on the latitude and longitude information of the regional automatic weather stations, find the grid points closest to the regional automatic weather stations from the high-resolution remote sensing inversion surface temperature grid data. The remote sensing temperature data value extracted from this grid point is Select 25 grid points centered on this grid point. The daily remote sensing data is obtained by averaging the data from 25 grid points, and is represented as follows:

[0097] ;

[0098] The remote sensing data sequence is composed of daily remote sensing data. , where m is the total number of years in the remote sensing data sequence, and l is the total number of days in the m-th year.

[0099] Scale separation is performed on remote sensing data to obtain remote sensing data anomaly sequences. Then, a regression model is constructed by combining it with the regional automatic station anomaly sequence, and the fitted anomaly sequence is calculated. , is represented as:

[0100] ;

[0101] Where n0 and n1 are the fitting coefficients; the total length of the automatic station anomaly sequence is l, then the fitting coefficients are:

[0102] ;

[0103] ;

[0104] The existing automatic station anomaly sequences (which contain missing data and have a relatively short time frame) are used as... Substituting the values ​​into the fitting coefficient formula, the fitting coefficient is calculated. Based on the fitting equation, the long-sequence remote sensing anomaly value is substituted to calculate the fitted long-sequence anomaly sequence. The time length of this fitted sequence is longer than that of the original regional automatic station anomaly sequence, and there are no missing measurements.

[0105] The scale separation of the observation data from the selected national standard meteorological stations described in step S3, and the scale separation of the remote sensing surface data described in step S4, adopt the same method as the scale separation of the regional automatic station observation data after quality control in step S2.

[0106] S5. Synthesize the fitted climatological sequence and the fitted anomaly sequence to obtain the reconstructed temperature sequence from the regional automatic weather stations, as follows:

[0107] ;

[0108] in, For the reconstruction sequence of regional automatic stations, To fit the anomaly sequence, To fit the climatological sequence; The number of days in a year, with a value from 1 to 366, where i represents the year.

[0109] The frequency distribution of the temperature sequence reconstructed from regional automatic weather stations and the temperature data reconstructed from regional automatic weather stations in this embodiment is as follows: Figure 3 and Figure 5As shown in the image. The spatial distribution map of daily maximum temperatures in Jiangsu Province during summer, constructed using temperature series reconstructed from regional automatic weather stations, is shown below. Figure 6 As shown.

[0110] Example 2: This example provides a non-transitory computer-readable storage medium storing computer instructions that cause a computer to execute a method for reconstructing regional automatic station temperature sequences based on scale separation and meteorological geographic zoning. The method includes the following steps:

[0111] S1. Select regional automatic weather stations within the study area whose observation data quality meets the requirements; and perform quality control on the observation data of the selected regional automatic weather stations.

[0112] S2. Scale separation is performed on the quality-controlled regional automatic station observation data to obtain the regional automatic station climatological sequence representing climatological scale temperature change, and the regional automatic station anomaly sequence obtained by subtracting the regional automatic station climatological sequence from the original temperature sequence.

[0113] S3. Based on climate-geographic zoning, select several national standard meteorological stations located in the same geographic zone as the regional automatic weather stations using the nearest distance method; perform scale separation on the observation data of the selected national standard meteorological stations to obtain the national station climatological sequence; then combine the regional automatic weather station climatological sequence and the national station climatological sequence to construct a linear multiple regression model and calculate the fitted climatological sequence.

[0114] S4. Based on the latitude and longitude information of the regional automatic weather station, find the grid point closest to the regional automatic weather station from the high-resolution remote sensing inversion surface temperature grid data. Using this grid point as the center, select grid data within a set range and average them to obtain remote sensing data. Perform scale separation on the remote sensing data to obtain the remote sensing data anomaly sequence. Combine the regional automatic weather station anomaly sequence and the surface data anomaly sequence to construct a linear regression model and calculate the fitted anomaly sequence.

[0115] S5. The fitted climatological sequence and the fitted anomaly sequence are combined to obtain the regional automatic station reconstruction sequence.

[0116] Example 3: This example provides an electronic device that may include a processor, a communication interface, a memory, and a communication bus. The processor, communication interface, and memory communicate with each other via the communication bus. The processor can call logical instructions from the memory to execute a regional automatic weather station temperature sequence reconstruction method based on scale separation and meteorological geographic zoning. This method includes the following steps:

[0117] S1. Select regional automatic weather stations within the study area whose observation data quality meets the requirements; and perform quality control on the observation data of the selected regional automatic weather stations.

[0118] S2. Scale separation is performed on the quality-controlled regional automatic station observation data to obtain the regional automatic station climatological sequence representing climatological scale temperature change, and the regional automatic station anomaly sequence obtained by subtracting the regional automatic station climatological sequence from the original temperature sequence.

[0119] S3. Based on climate-geographic zoning, select several national standard meteorological stations located in the same geographic zone as the regional automatic weather stations using the nearest distance method; perform scale separation on the observation data of the selected national standard meteorological stations to obtain the national station climatological sequence; then combine the regional automatic weather station climatological sequence and the national station climatological sequence to construct a linear multiple regression model and calculate the fitted climatological sequence.

[0120] S4. Based on the latitude and longitude information of the regional automatic weather station, find the grid point closest to the regional automatic weather station from the high-resolution remote sensing inversion surface temperature grid data. Using this grid point as the center, select grid data within a set range and average them to obtain remote sensing data. Perform scale separation on the remote sensing data to obtain the remote sensing data anomaly sequence. Combine the regional automatic weather station anomaly sequence and the surface data anomaly sequence to construct a linear regression model and calculate the fitted anomaly sequence.

[0121] S5. The fitted climatological sequence and the fitted anomaly sequence are combined to obtain the regional automatic station reconstruction sequence.

[0122] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and 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.

[0123] Example 4: This example provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can perform a method for reconstructing regional automatic station temperature sequences based on scale separation and meteorological geographic zoning. This method includes the following steps:

[0124] S1. Select regional automatic weather stations within the study area whose observation data quality meets the requirements; and perform quality control on the observation data of the selected regional automatic weather stations.

[0125] S2. Scale separation is performed on the quality-controlled regional automatic station observation data to obtain the regional automatic station climatological sequence representing climatological scale temperature change, and the regional automatic station anomaly sequence obtained by subtracting the regional automatic station climatological sequence from the original temperature sequence.

[0126] S3. Based on climate-geographic zoning, select several national standard meteorological stations located in the same geographic zone as the regional automatic weather stations using the nearest distance method; perform scale separation on the observation data of the selected national standard meteorological stations to obtain the national station climatological sequence; then combine the regional automatic weather station climatological sequence and the national station climatological sequence to construct a linear multiple regression model and calculate the fitted climatological sequence.

[0127] S4. Based on the latitude and longitude information of the regional automatic weather station, find the grid point closest to the regional automatic weather station from the high-resolution remote sensing inversion surface temperature grid data. Using this grid point as the center, select grid data within a set range and average them to obtain remote sensing data. Perform scale separation on the remote sensing data to obtain the remote sensing data anomaly sequence. Combine the regional automatic weather station anomaly sequence and the surface data anomaly sequence to construct a linear regression model and calculate the fitted anomaly sequence.

[0128] S5. The fitted climatological sequence and the fitted anomaly sequence are combined to obtain the regional automatic station reconstruction sequence.

[0129] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, 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 according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0130] 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.

[0131] 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 reconstructing regional automatic weather station temperature sequences based on scale separation and meteorological geographic zoning, characterized in that, Includes the following steps: S1. Select regional automatic weather stations within the study area whose observation data quality meets the requirements; and perform quality control on the observation data of the selected regional automatic weather stations. S2. Scale separation is performed on the quality-controlled regional automatic station observation data to obtain the regional automatic station climatological sequence representing climatological scale temperature change, and the regional automatic station anomaly sequence obtained by subtracting the regional automatic station climatological sequence from the original temperature sequence. S3. Based on climate-geographic zoning, select several national standard meteorological stations located in the same geographic zone as the regional automatic weather stations using the nearest distance method; perform scale separation on the observation data of the selected national standard meteorological stations to obtain the national station climatological sequence; then combine the regional automatic weather station climatological sequence and the national station climatological sequence to construct a linear multiple regression model and calculate the fitted climatological sequence. S4. Based on the latitude and longitude information of the regional automatic weather station, find the grid point closest to the regional automatic weather station from the high-resolution remote sensing inversion surface temperature grid data. Using this grid point as the center, select grid data within a set range and average them to obtain remote sensing data. Perform scale separation on the remote sensing data to obtain the remote sensing data anomaly sequence. Combine the regional automatic weather station anomaly sequence and the surface data anomaly sequence to construct a linear regression model and calculate the fitted anomaly sequence. S5. The fitted climatological sequence and the fitted anomaly sequence are combined to obtain the regional automatic station reconstruction sequence.

2. The method for reconstructing regional automatic weather station temperature sequences based on scale separation and meteorological geographic zoning according to claim 1, characterized in that, Step S1 involves selecting regional automatic weather stations within the study area whose observation data quality meets the requirements; specifically: Extract the construction information of regional automatic weather stations within the study area, determine the station locations and surrounding environment, and remove regional automatic weather stations used for traffic observation and large water body observation; and perform observation quantity statistics on the observation data of the remaining regional automatic weather stations, and remove regional automatic weather stations with data duration of less than 5 years and missing data exceeding 45% of the total data volume.

3. The method for reconstructing regional automatic weather station temperature sequences based on scale separation and meteorological geographic zoning according to claim 1, characterized in that, Step S1 involves quality control of the selected area's automatic weather station observation data; specifically, this includes: (1) Temperature threshold check: Determine the upper and lower thresholds of annual temperature based on the historical temperature observations of the study area, and remove data from the automatic station observation data of the area that exceed the upper and lower thresholds of annual temperature; (2) Statistically analyze the observation data of the national standard meteorological stations around the regional automatic stations, calculate the monthly temperature upper and lower thresholds, and remove the data in the regional automatic station observation data that exceed the monthly temperature upper and lower thresholds for the corresponding month; (3) Based on the observation data of the national standard meteorological station, calculate the temperature change threshold for 24, 48 and 72 hours in the area where the regional automatic station is located, and remove the data in the regional automatic station observation data that exceed the temperature change threshold; (4) Set the standard deviation of temperature change within a specified time period and remove data from the automatic station observation data of the area that exceed ±3 times the standard deviation within that time period.

4. The method for reconstructing regional automatic weather station temperature sequences based on scale separation and meteorological geographic zoning according to claim 1, characterized in that, Step S2 describes the scale separation of the quality-controlled regional automatic weather station observation data to obtain the regional automatic weather station climatological sequence representing climatological-scale temperature changes, and the regional automatic weather station anomaly sequence obtained by subtracting the regional automatic weather station climatological sequence from the original temperature sequence; specifically: S21, Assume there are a total of regional automatic stations. Year data, number The original temperature sequence for the year is ,calculate Annual daily climate average series: ； S22. Using the moving average method, a moving average is calculated on the daily climate average series, with a moving window of 31 days and a moving average of 1 day at a time. and By performing a cyclic sliding process, the regional automatic weather station climatological sequence is obtained, as follows: ； This is a climatological sequence from regional automatic weather stations; S23. Subtract the regional automatic station climatological sequence from the original temperature sequence of the regional automatic stations to obtain the regional automatic station anomaly sequence. , represented as: ； This is an automatic station distance sequence for the region.

5. The method for reconstructing regional automatic weather station temperature sequences based on scale separation and meteorological geographic zoning according to claim 4, characterized in that, The scale separation of the observation data from the selected national standard meteorological station described in step S3, and the scale separation of the remote sensing data described in step S4, adopt the same method as the scale separation of the regional automatic station observation data after quality control in step S2.

6. The method for reconstructing regional automatic weather station temperature sequences based on scale separation and meteorological geographic zoning according to claim 5, characterized in that, In step S3, scale separation is performed on the observation data of the selected national standard meteorological stations to obtain the national station climatological series. Then, a linear multiple regression model was constructed using it and the climatological sequences from regional automatic weather stations to calculate the fitted climatological sequences. , represented as: ； in, This is the national station climatological sequence of the k-th national standard meteorological station; is the regression coefficient.

7. The method for reconstructing regional automatic weather station temperature sequences based on scale separation and meteorological geographic zoning according to claim 5, characterized in that, Step S4 involves finding the grid point closest to the regional automatic weather station from the high-resolution remote sensing inversion surface temperature grid data based on the latitude and longitude information of the regional automatic weather station, and then averaging the grid data within a set range centered on that grid point to obtain remote sensing data; specifically: Based on the latitude and longitude information of the regional automatic weather stations, the grid points closest to the regional automatic weather stations are found from the high-resolution remote sensing inversion grid data of land surface temperature. The remote sensing temperature data value extracted from this grid point is ; Select 25 grid points centered on this grid point. The daily remote sensing data is obtained by averaging the data from 25 grid points, and is represented as follows: ； The remote sensing data sequence is composed of daily remote sensing data. , where m is the total number of years in the remote sensing data sequence, and l is the total number of days in the m-th year.

8. The method for reconstructing regional automatic weather station temperature sequences based on scale separation and meteorological geographic zoning according to claim 7, characterized in that, In step S4, scale separation is performed on the remote sensing data sequence to obtain the remote sensing data anomaly sequence. Then, a regression model is constructed by combining it with the regional automatic station anomaly sequence, and the fitted anomaly sequence is calculated. , represented as: ； Where n0 and n1 are the fitting coefficients.

9. The method for reconstructing regional automatic weather station temperature sequences based on scale separation and meteorological geographic zoning according to claim 7, characterized in that, Step S5 involves synthesizing the fitted climatological sequence and the fitted anomaly sequence to obtain the regional automatic weather station reconstruction sequence, as follows: ； in, For the reconstruction sequence of regional automatic stations, To fit the anomaly sequence, To fit the climatological sequence; The number of days in a year, with a value from 1 to 366, where i represents the year.

10. The method for reconstructing regional automatic weather station temperature sequences based on scale separation and meteorological geographic zoning according to any one of claims 1-9, characterized in that, In the calculations of steps S2 to S5, each year is counted as 366 days as a leap year. If the number of observations on a certain date is less than 50% of the total number of observations for the year, the observation data for that date is recorded as missing, and all data for that date are not included in the calculation.

Images