A method for quality control of multi-factor coordinated meteorological data of frozen ground, snow depth, and condensation weather phenomena in Xinjiang

Abstract

The present invention discloses a method for quality control of multi-factor coordinated meteorological data of frozen soil, snow depth, and condensation-type weather phenomena in Xinjiang, comprising the following steps: obtaining full-factor meteorological data and monthly observation data of daily frozen soil, snow depth, and condensation-type weather phenomena from national ground meteorological observation stations in Xinjiang; constructing relationship expressions between frozen soil, snow depth, and condensation-type weather phenomena and meteorological elements respectively, and obtaining theoretical frozen soil depth, theoretical snow depth, and theoretical condensation-type weather phenomena; comparing and analyzing the actual frozen soil depth, actual snow depth, and actual condensation-type weather phenomena in the monthly observation data with the theoretical frozen soil depth, theoretical snow depth, and theoretical condensation-type weather phenomena, screening out suspicious data, and judging the rationality of the elements of frozen soil, snow depth, and condensation-type weather phenomena. The method for quality control of multi-factor coordinated meteorological data of frozen soil, snow depth, and condensation-type weather phenomena is used, and the coordinated method is applied to the monthly provincial quality control work to make the meteorological data more accurate.

Description

Technical Field

[0001] The present invention relates to the field of meteorological ground data processing application technology, and more specifically to a method for controlling meteorological data quality in coordination with multiple factors of frozen soil, snow depth, and condensation weather phenomena in Xinjiang. Background Art

[0002] Frozen soil is directly related to our lives. In urban construction, the depth of subway construction, water pipelines, and oil pipelines are all related to the depth of frozen soil. According to the depth data of frozen soil, relevant environmental conditions are evaluated to predict whether the site is suitable for construction. Frozen soil observation is an important data source for evaluating construction conditions in plateaus and cold regions. It can measure 0-2.5m underground. However, its shortcomings are that it is limited to measurements at national meteorological observation stations, the detection range of frozen soil measurements is limited, and human factors have a greater impact.

[0003] In my country, since 2016, national observatories (and some regional stations) throughout Xinjiang have successively installed automatic snow depth instruments, and automatic permafrost equipment has also entered the stage of parallel observation with manual comparison. In 2020, the National Meteorological Administration optimized and improved the observation items and methods, cancelled the manual observation of weather phenomena such as condensation frost, ice and snow accumulation, and adopted automatic comprehensive identification to carry out observations. The improvement of meteorological observation methods has generated a large amount of intensive observation data. However, the quality control and audit rules of the business system in the current business work are extremely imperfect, and data omissions and anomalies cannot be discovered in time. They can only be discovered through factor correlation checks, time consistency checks, and integrity checks when the monthly data files are manually reviewed in the next month. There are timeliness issues. At present, the quality control methods of permafrost data and ground temperature, snow depth and condensation weather phenomena are not closely coordinated. Permafrost data are often missed, and there are multiple observation elements in snow depth, frost and ice accumulation, which cannot be discovered by the current business system quality control technology.

[0004] Therefore, in response to the above specific problems, it is urgent to develop a more reasonable, comprehensive and effective method for coordinated meteorological data quality control of multiple factors such as frozen ground, snow depth and condensation weather phenomena to overcome the above defects and shorten the time limit for meteorological data quality control. Summary of the invention

[0005] To this end, the technical problem to be solved by the present invention is to provide a multi-factor coordinated meteorological data quality control method for frozen ground, snow depth, and condensation weather phenomena in Xinjiang, so as to solve the problems of missed measurements, multiple measurements, failure to promptly discover suspected erroneous data, and poor timeliness of frozen ground, snow depth, and condensation weather phenomena.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] A method for controlling the quality of multi-factor coordinated meteorological data of frozen ground, snow depth, and condensation weather phenomena in Xinjiang, comprising the following steps:

[0008] Step P1: Obtain full-factor meteorological data and monthly observation data of daily frozen ground, snow depth, and condensation weather phenomena from national ground-based meteorological observation stations in Xinjiang;

[0009] Step P2: construct the relationship expressions between frozen soil, snow depth, condensation-type weather phenomena and meteorological elements respectively, and obtain the theoretical frozen soil depth, theoretical snow depth and theoretical condensation-type weather phenomena;

[0010] Step P3: Compare and analyze the actual frozen soil depth, actual snow depth and actual condensation-type weather phenomena in the monthly observation data in step P1 with the theoretical frozen soil depth, theoretical snow depth and theoretical condensation-type weather phenomena in step P2, filter out suspicious data, and determine the rationality of the frozen soil, snow depth and condensation-type weather phenomenon elements.

[0011] In the above-mentioned method for quality control of multi-factor coordinated meteorological data of frozen ground, snow depth and condensation weather phenomena in Xinjiang, in step P1,

[0012] 105 national meteorological stations with complete observation elements were selected from all meteorological stations in Xinjiang. The standard was that the obtained observation data had a complete sequence. At the same time, the selection of observation stations could reflect the local terrain, landform and climate characteristics.

[0013] The above-mentioned method for multi-factor coordinated meteorological data quality control of frozen ground, snow depth, and condensation weather phenomena in Xinjiang region, step P1 includes the following:

[0014] Step P101: Obtain full-factor meteorological observation monthly data from 105 national ground stations across Xinjiang;

[0015] Step P102: Find the contents of the A, Z, W, and DB fields in the full-factor monthly meteorological observation data of 105 national ground stations across Xinjiang, and obtain daily observation data on frozen ground, snow depth, and condensation weather phenomena.

[0016] In the above-mentioned method for quality control of meteorological data of multi-factor coordinated frozen soil, snow depth, and condensation weather phenomena in Xinjiang, in step P2, based on the significance analysis of the meteorological elements of frozen soil and weather phenomena, it is found that there is a significant correlation between frozen soil and the corresponding ground temperature of 0 cm, ground temperature of 5 cm, and ground temperature of 15 cm, thereby constructing a theoretical frozen soil depth and ground temperature correlation expression:

[0017] f(T t )=aT DB And a<0, T DB ≤0 (1);

[0018] Among them, f(T t ) is the theoretical frozen soil depth, T DB is ground temperature, a is f(T t ) and T DB The correlation coefficient is negative.

[0019] In the above-mentioned method for quality control of meteorological data of multi-factor coordinated meteorological phenomena of frozen soil, snow depth and condensation in Xinjiang, in step P2, based on the significance analysis of snow depth and meteorological elements of weather phenomena, it is found that there is a significant correlation between snow depth and snowfall weather phenomena, existing snow depth and ground temperature, thereby constructing a correlation expression between theoretical snow depth and snowfall weather phenomena, existing snow depth and ground temperature:

[0020] f(z t )≥0, and:

[0021] When T DB ≥0, then f(z t )=cW+Z+bT DB , c>0, W≥0, Z≥0, b<0 (2);

[0022] When T DB When <0, then f(z t )=cW+Z, c>0, W≥0, Z≥0 (3);

[0023] Among them, f(z t ) is the theoretical snow depth; W is the weather phenomenon of snow, and W ≥ 0; T DB is ground temperature; Z is the existing snow depth, and Z>0; b is f(z t ) and T DB The correlation coefficient of , and b<0, that is, f(z t ) and T DB Negatively correlated; c is f(z t ) and W, and c>0, that is, f(z t ) is positively correlated with c.

[0024] In the above-mentioned method for quality control of meteorological data of multi-factor coordinated meteorological phenomena of frozen soil, snow depth and condensation in Xinjiang, in step P2, condensation weather phenomena include frost, ice and snow weather phenomena. Based on the significance analysis of condensation weather and meteorological elements of weather phenomena, it is found that there is a significant correlation between condensation weather and ground temperature. The correlation between condensation weather and ground temperature T DB Negatively correlated, that is, when T DB When <0, condensation-like weather phenomena will theoretically occur.

[0025] The above-mentioned method for multi-factor coordinated meteorological data quality control of frozen soil, snow depth, and condensation weather phenomena in Xinjiang region includes, in step P3:

[0026] Step P301: Perform data quality control on all-element meteorological data from 105 national ground stations, so that each meteorological element passes internal consistency check, time consistency check, climate limit value and spatial consistency check;

[0027] Step P302: Obtain the actual permafrost depth, actual snow depth and actual condensation-like weather phenomena in the monthly observation data of Xinjiang on that day, compare them with the theoretical permafrost depth, theoretical snow depth and theoretical condensation-like weather phenomena of Xinjiang on that day, mark the suspected erroneous data, use the relevant data of the previous day and the next day to screen and correct the suspected erroneous data, and verify the authenticity of the data.

[0028] The above-mentioned method for quality control of meteorological data of multi-factor coordination of frozen soil, snow depth and condensation weather phenomena in Xinjiang is aimed at frozen soil data.

[0029] Step P401-1: Filter out the frozen soil data in the A field and the ground temperature data in the DB field in the monthly data file, and use the frozen soil data in the A field to obtain the actual frozen soil depth data A. t At the same time, using the ground temperature data in the DB field, the theoretical frozen soil depth f(T t );

[0030] Step P401-2: The actual frozen soil depth data A t and theoretical frozen soil depth f(T t ) Compare the data of these two arrays and traverse A t (t=1,2,···,n) frozen soil observation series data and theoretical frozen soil depth f(T t )(t=1,2,···,n) t ), we get formula (4):

[0031] f(A t )=A t -f(T t ) (4);

[0032] Step P401-3: Screen out f(A t ) is a negative value, and the actual frozen soil depth data A t Zero or missing, ground temperature 5cm T DB Data <0 are marked as suspicious frozen soil data;

[0033] Step P401-4: Compare the selected suspicious frozen soil data with the difference f(A t-1 ) and the difference f(A t+1 ), check the suspicious data f(A t ) is true, if [f(A t)< <f(A t-1 )]∪[f(A t )< <f(A t+1 )] value, it is suspected erroneous data, otherwise it is correct data.

[0034] The above-mentioned method for quality control of meteorological data of multi-factor coordination of frozen ground, snow depth and condensation weather phenomena in Xinjiang is aimed at snow depth data.

[0035] Step P402-1: Filter out the existing snow depth data in the Z field, the snowy weather phenomenon data in the W field, and the ground temperature data in the DB field in the monthly data file;

[0036] Step P402-2: Establish actual snow depth data Z t and theoretical snow depth f(z t ) These two arrays: the actual snow depth data Z is obtained from the existing snow depth data in the Z field and the snowy weather phenomenon data in the W field t ; Based on the ground temperature data in the DB field, the theoretical snow depth f(z t ) data; at the same time, the actual snow depth data Z t and theoretical snow depth f(z t ) to process the data, which is divided into two cases: snow depth and no snow depth. Snow depth is represented as 1, and no snow depth is represented as 0;

[0037] Step P402-3: Compare the data of the two arrays and traverse Z t (t=1,2,···,n) Actual snow depth observation sequence data and f(z t )(t=1,2,···,n) theoretical snow depth data, and obtain the difference between the two data sets f(Z t ), we get formula (5):

[0038] f(Z t )=Z t -f(z t ) (5);

[0039] Step P402-3: If the snow depth difference f(Z t ) data is 1 or -1, it is marked as suspicious snow depth data;

[0040] Step P402-4: For the suspicious snow depth data that has been screened out, use the ground temperature data and snow depth data Z of the previous day to t-1 and the ground temperature data and snow depth data of the next day Z t+1 Screen the snow depth and monthly meteorological observation data to verify the authenticity of the suspected erroneous data:

[0041] When the marked ground temperature is greater than zero degrees, there is suddenly a deep snow

[0042] T t-1 >0,Z t-1 =0)∩(T t >0,Z t =1)∩(T t+1 >0,Z t+1 =0), the ground temperature is below zero and there is no snow depth (T t-1 <0,Z t-1 =1)∩(T t <0,Z t =0)∩(T t+1 <0,Z t+1 =1), and combined with the weather phenomenon of the day, the "70" field marks that there is no snow and suddenly there is snow depth W t =0,Z t =1. When it snows suddenly, the snow depth is not deep. t >0,Z t =0, it is suspected erroneous data, otherwise it is correct data.

[0043] The above-mentioned method for multi-factor coordinated meteorological data quality control of frozen ground, snow depth, and condensation weather phenomena in Xinjiang is aimed at condensation weather phenomena:

[0044] Step P403-1: Filter out the weather phenomenon data in the W field and the ground temperature data in the DB field in the monthly data file;

[0045] Step P403-2: Data processing: the fields with "02" frost, "03" ice and "16" snow in the weather phenomenon data of the W field are marked as W. t The corresponding DB field ground temperature data is marked as T t ;

[0046] Step P403-3: Create condensation weather data W t , corresponding to ground temperature T t These two arrays; get formula (6):

[0047] f(W)=W t , T t >0 (6);

[0048] Step P403-4: Using W t and T t Check the authenticity of suspicious data. t >0, and the weather data W t Data recorded as “02” frost, “03” ice and “16” snow accumulation are suspected to be erroneous data.

[0049] The technical solution of the present invention achieves the following beneficial technical effects:

[0050] 1. Utilize the collaborative quality control method of multiple factors such as frozen soil, snow depth, and condensation weather phenomena, apply this collaborative method to the monthly provincial quality control work, and provide cause analysis and improvement measures for months with more erroneous data, so as to make meteorological data more accurate.

[0051] 2. Theoretical values ​​are obtained with the help of the accuracy of conventional meteorological elements such as ground temperature of automatic weather stations, and actual values ​​are obtained by recording data from observation stations with similar locations. The theoretical values ​​are compared with the actual values, and filtering conditions are set according to the data recorded by the observation stations to filter out missed and erroneous data in the automatic observation data, and verify the effectiveness of the algorithm.

[0052] 3. We analyzed the causes of the months with more erroneous automatic observation data and provided preliminary improvement measures.

[0053] Regarding the depth of permafrost, there are many abnormal data in the early stage of permafrost freezing in November and the thawing period in March of the following year. The values ​​of the automatic permafrost sensor (trial) at the same observation time can be used as a reference to reduce erroneous data; Regarding the snow depth data, there are many abnormal data in the early stage of winter snowfall in October and the late stage of winter snowfall in February of the following year. Professional and technical personnel can be organized to check the snow depth data of the entire Xinjiang every day in combination with the current meteorological data platform; Regarding condensation weather phenomena, there are many abnormal data because the meteorological station personnel are not familiar with the business regulations for recording ice and frost, which requires special attention.

[0054] 4. Xinjiang is located at a high altitude, with a long winter, heavy snowfall and deep permafrost, which makes it difficult to develop and utilize Xinjiang's rich underground mineral resources.

[0055] The multi-factor coordinated meteorological data quality control method for frozen ground, snow depth, and condensation weather phenomena proposed in this application is of great value to mineral development and resource utilization in Xinjiang, and has the advantages of simple method, easy calculation, simplicity and practicality.

[0056] 5. Since 2016, automatic snow depth instruments and frozen soil automatic equipment have been installed one after another. The frozen soil observation and snow depth observation in China have entered the parallel observation stage. There are still large differences between the automatic frozen soil data and the artificial frozen soil observation values, as well as the automatic snow depth data and the artificial snow depth observation values. This will greatly increase the scope of application of this application algorithm. For example, quality control can be performed on hundreds of automatic and manual stations respectively to reduce errors caused by instrument performance and human interference factors and improve data quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 Elevation map of northern Xinjiang, southern Xinjiang and eastern Xinjiang in some areas of Xinjiang in the embodiment of the present invention;

[0058] Figure 2 Comparison diagram of the frozen soil and ground temperature quality control in northern Xinjiang before and after the embodiment of the present invention;

[0059] Figure 3 Comparison diagram of frozen soil and ground temperature quality control in East Xinjiang before and after the embodiment of the present invention;

[0060] Figure 4 Comparison diagram of frozen soil and ground temperature quality control in southern Xinjiang before and after the embodiment of the present invention;

[0061] Figure 5 Comparison diagram of frozen soil and ground temperature quality control before and after at an altitude of 1000 meters or more in the embodiment of the present invention;

[0062] Figure 6 Comparison diagram of correlation coefficients between frozen soil and ground temperature in northern Xinjiang before and after quality control in the embodiment of the present invention;

[0063] Figure 7 Comparison diagram of correlation coefficients between frozen soil and ground temperature in East Xinjiang before and after quality control in the embodiment of the present invention;

[0064] Figure 8 Comparison diagram of correlation coefficients between frozen soil and ground temperature in southern Xinjiang before and after quality control in the embodiment of the present invention;

[0065] Fig. 9 Comparison of correlation coefficients before and after quality control of frozen soil and ground temperature above 1000 meters above sea level in the embodiment of the present invention;

[0066] Fig.10 Comparison diagram of snow depth and ground temperature in northern Xinjiang before and after quality control in an embodiment of the present invention;

[0067] Fig.11 Comparison diagram of ice and frost quality control before and after in Beijiang in the embodiment of the present invention;

[0068] Fig.12 Comparison diagram of the correlation coefficient between the snow depth in Dongjiang and the ground temperature before and after quality control in the embodiment of the present invention. DETAILED DESCRIPTION

[0069] 1. Materials and methods

[0070] 1.1 Research Materials

[0071] The data on frozen ground, snow depth, and condensation weather phenomena used in this implementation are all from the Xinjiang Uygur Autonomous Region Meteorological Information Center (Xinjiang Uygur Autonomous Region Meteorological Archives). The data were strictly quality controlled during the production process, and suspicious and erroneous data were manually checked and corrected. The statistical data were obtained in strict accordance with the statistical regulations and methods in the "Ground Meteorological Observation Specifications (2003 Edition)", and there are detailed metadata information and quality control information. The data were processed as monthly averages during the data collation process.

[0072] 105 national meteorological stations with complete observation elements were selected from all meteorological stations in Xinjiang. The standard was that the obtained observation data had a complete sequence. At the same time, the selection of observation stations could reflect the local terrain, landform and climate characteristics.

[0073] In this embodiment, based on Xinjiang's geographical environment, topography, altitude, underlying surface conditions, climate zoning and other factors, Xinjiang is divided into northern Xinjiang, southern Xinjiang, eastern Xinjiang, and stations above 1,000 meters above sea level. The monthly average surface temperature, monthly minimum surface temperature, and monthly average 5cm-40cm ground temperature data of 20 stations out of 105 national meteorological stations are selected for research. The time period is selected from October 1, 2023 to March 31, 2024. The observation instruments are unified as artificial permafrost devices issued by the Xinjiang Meteorological Bureau to ensure the uniformity and scientificity of the quality control process.

[0074] 1.2 Research Methods

[0075] like Figure 1 As shown, the quality control method in this embodiment specifically includes the areas in Xinjiang, namely, northern Xinjiang, southern Xinjiang, eastern Xinjiang, and areas above 1,000 meters above sea level. Taking eastern Xinjiang as an example, the specific steps are as follows:

[0076] Step P1, select the monthly observation data of Hami, Naomaohu, Shanshan and Turpan national ground stations from the meteorological observation stations in the eastern Xinjiang region;

[0077] The main principles for selecting quality control data observation stations are: based on the geographical location, topography, altitude, and underlying surface of the selected observation stations, they are divided into east, north, south, and west regions within the East Xinjiang area. The eastern part of East Xinjiang is Hami Station, the northern part is Naomao Lake Station, the southern part is Shanshan Station, and the western part is Turpan Station.

[0078] In this embodiment, the data period is the frozen soil freezing period from October 1, 2023 to March 31, 2024. The target site is selected to manually observe the ground meteorological data at 08:00, and the frozen soil observation instruments are uniformly issued by the Xinjiang Observation Network Department. The artificial frozen soil reading is 0.1cm.

[0079] Daily observations of frozen ground, snow depth, and condensation-type weather phenomena were obtained from monthly data from various national ground stations in East Xinjiang;

[0080] Step P101: Obtain full-factor meteorological observation monthly data from 24 national ground stations at four stations in East Xinjiang; perform data quality control on the acquired full-factor meteorological data, so that each meteorological element passes the internal consistency check, time consistency check, climate limit value, and spatial consistency check;

[0081] Step P202: Find the contents of the A, Z, W, and DB fields in the full-factor meteorological observation monthly data of the four national ground stations in East Xinjiang, and obtain daily observation data on frozen ground, snow depth, and condensation weather phenomena.

[0082] Step P2: Correlation analysis is performed on the observation data of frozen soil, snow depth, and condensation weather phenomena with the corresponding meteorological elements such as daily temperature, ground temperature, and weather phenomena.

[0083] For frozen soil: Based on the significance analysis of frozen soil and meteorological elements of weather phenomena, it is found that there is a significant correlation between frozen soil and the corresponding daily air temperature, ground temperature 0cm, ground temperature 5cm, and ground temperature 15cm, thus constructing the theoretical frozen soil depth and ground temperature correlation expression:

[0084] f(T t )=aT DB And a<0, T DB ≤0(1);

[0085] Among them, f(T t ) is the theoretical frozen soil depth, T DB is ground temperature, a is f(T t ) and T DB The correlation coefficient is negative.

[0086] like Figure 6 , Figure 7 , Figure 8 , Fig. 9 The changes in the correlation coefficient a of the surface temperature, ground temperature of 5cm, ground temperature of 10cm, ground temperature of 15cm, ground temperature of 20cm and ground temperature of 40cm in northern Xinjiang, eastern Xinjiang, southern Xinjiang and above 1000 meters above sea level before and after the internal consistency check, time consistency check, climate boundary value and spatial consistency check.

[0087] Regarding snow depth: Based on the significance analysis of snow depth and meteorological elements of weather phenomena, it is found that there is a significant correlation between snow depth and snowfall weather phenomena, existing snow depth, and ground temperature, thus constructing a correlation expression between theoretical snow depth and snowfall weather phenomena, existing snow depth, and ground temperature:

[0088] f(z t )≥0, and:

[0089] When T DB ≥0, then f(z t )=cW+Z+bT DB , c>0, W≥0, Z≥0, b<0 (2);

[0090] When T DB When <0, then f(z t )=cW+Z, c>0, W≥0, Z≥0 (3);

[0091] Among them, f(z t ) is the theoretical snow depth; W is the weather phenomenon of snow, and W ≥ 0; T DB is ground temperature; Z is the existing snow depth, and Z>0; b is f(z t ) and T DB The correlation coefficient of f(z) and T DB Negatively correlated; c is f(z t ) and W, and c>0, that is, f(z t ) is positively correlated with c.

[0092] like Fig.12 Shown is the change in correlation coefficient b between snow depth and surface temperature, ground temperature 5 cm, ground temperature 10 cm, ground temperature 15 cm, ground temperature 20 cm and ground temperature 40 cm in Dongjiang before and after internal consistency check, time consistency check, climate boundary value and spatial consistency check.

[0093] For condensation weather phenomena:

[0094] Condensation weather phenomena include frost, ice and snow. Based on the significance analysis of condensation weather and meteorological elements of weather phenomena, it is found that there is a significant correlation between condensation weather and ground temperature. DB Negatively correlated, that is, when T DB When <0, condensation-like weather phenomena will theoretically occur.

[0095] Step P3: Analyze the suspected erroneous data in the observation sequence of frozen ground, snow depth, and condensation weather phenomena through data integrity, climate limit values, and element correlation checks, and determine the rationality of the elements of frozen ground, snow depth, and condensation weather phenomena.

[0096] Step P301: Perform data quality control on all-element meteorological data of national ground stations, so that each meteorological element passes internal consistency check, time consistency check, climate limit value and spatial consistency check;

[0097] Step P302: Obtain the actual permafrost depth, actual snow depth and actual condensation-like weather phenomena in the monthly observation data of Xinjiang on that day, compare them with the theoretical permafrost depth, theoretical snow depth and theoretical condensation-like weather phenomena of Xinjiang on that day, mark the suspected erroneous data, use the relevant data of the previous day and the next day to screen and correct the suspected erroneous data, and verify the authenticity of the data.

[0098] For frozen soil data,

[0099] Step P401-1: Filter out the frozen soil data in the A field and the ground temperature data in the DB field in the monthly data file, and use the frozen soil data in the A field to obtain the actual frozen soil depth data A. t At the same time, using the ground temperature data in the DB field, the theoretical frozen soil depth f(T t );

[0100] Step P401-2: The actual frozen soil depth data A t and theoretical frozen soil depth f(T t ) Compare the data of these two arrays and traverse A t (t=1,2,···,n) frozen soil observation series data and theoretical frozen soil depth f(T t )(t=1,2,···,n) t ), we get formula (4):

[0101] f(A t )=A t -f(T t ) (4);

[0102] Step P401-3: Screen out f(A t ) is a negative value, and the actual frozen soil depth data A t Zero or missing, ground temperature 5cm T DB Data <0 are marked as suspicious frozen soil data;

[0103] Step P401-4: Compare the selected suspicious frozen soil data with the difference f(A t-1 ) and the difference f(A t+1 ), check the suspicious data f(A t ) is true, if [f(A t )< <f(A t-1 )]∪[f(A t )< <f(A t+1 )] value, it is suspected erroneous data, otherwise it is correct data.

[0104] like Figure 2 (Northern Xinjiang), Figure 3 (Eastern Xinjiang), Figure 4The comparison chart of frozen soil and ground temperature before and after quality control in (Southern Xinjiang) shows that the situation in northern Xinjiang is similar to that in eastern Xinjiang. The frozen soil depth before and after quality control of frozen soil and ground temperature in October, December, January and February is the same, but in November and March of the following year, there are erroneous data. The frozen soil depth after quality control in November is significantly greater than the frozen soil depth before quality control, and the frozen soil depth after quality control in March of the following year is significantly less than the frozen soil depth before quality control. November is the early stage of frozen soil freezing, and March is the thawing period. There will be manual recording errors in the early stage of freezing and the thawing period. In these two periods, the values ​​of the automatic frozen soil sensor (trial) at the same observation time can be used as a reference.

[0105] Figure 5 This is a comparison chart before and after quality control of permafrost and ground temperature above 1,000 meters above sea level. Analyzing the data of stations above 1,000 meters above sea level from a spatial level, it can be seen that the time when erroneous data appeared in northern Xinjiang, eastern Xinjiang and southern Xinjiang above 1,000 meters above sea level is the same, all in November and March of the following year. This is still related to the unstable freezing period, but it can be seen that due to the high altitude and low ground temperature, the erroneous values ​​are significantly reduced.

[0106] Special attention is paid to the permafrost conditions during the freeze-thaw period in November and March, so as to obtain accurate data on permafrost depth and provide assistance for permafrost construction operations in Xinjiang.

[0107] Step P402-1: Filter out the existing snow depth data in the Z field, the snowy weather phenomenon data in the W field, and the ground temperature data in the DB field in the monthly data file. For the snow depth data, Fig.10 ;

[0108] Step P402-2: Establish actual snow depth data Z t and theoretical snow depth f(z t ) These two arrays: the actual snow depth data Z is obtained from the existing snow depth data in the Z field and the snowy weather phenomenon data in the W field t ; Based on the ground temperature data in the DB field, the theoretical snow depth f(z t ) data; at the same time, the actual snow depth data Z t and theoretical snow depth f(z t ) to process the data, which is divided into two cases: snow depth and no snow depth. Snow depth is represented as 1, and no snow depth is represented as 0;

[0109] Step P402-3: Compare the data of the two arrays and traverse Z t (t=1,2,···,n) Actual snow depth observation sequence data and f(z t )(t=1,2,···,n) theoretical snow depth data, and obtain the difference between the two data sets f(Z t ), we get formula (5):

[0110] f((Z t )=Z t -f(z t ) (5);

[0111] Step P402-3: If the snow depth difference f(Z t ) data is 1 or -1, it is marked as suspicious snow depth data;

[0112] Step P402-4: For the suspicious snow depth data that has been screened out, use the ground temperature data and snow depth data Z of the previous day to t-1 and the ground temperature data and snow depth data of the next day Z t+1 Screen the snow depth and monthly meteorological observation data to verify the authenticity of the suspected erroneous data:

[0113] When the marked ground temperature is greater than zero degrees, there is a sudden snow depth T t-1 >0,Z t-1 =0)∩(T t >0,Z t =1)∩(T t+1 >0,Z t+1 =0), the ground temperature is below zero and there is no snow depth (T t-1 <0,Z t-1 =1)∩(T t <0,Z t =0)∩(T t+1 <0,Z t+1 =1; and combined with the weather phenomenon of the day, the "70" field marks that there is no snow and suddenly there is snow depth (W t =0,Z t =1) When the snow suddenly becomes shallow (W t >0,Z t =0), it is suspected erroneous data, otherwise it is correct data.

[0114] like Fig.10 As shown, there are suspected erroneous data in October and February of the following year. October is the beginning of winter snowfall, and February is the end of winter snowfall. There will be manual recording errors in the early and late snowfall periods. During these two periods, professional and technical personnel can be organized to check the snow depth data in Xinjiang every day in combination with the current meteorological data platform.

[0115] For condensation weather phenomena:

[0116] Step P403-1: Filter out the weather phenomenon data in the W field and the ground temperature data in the DB field in the monthly data file;

[0117] Step P403-2: Data processing: the fields with "02" frost, "03" ice and "16" snow in the weather phenomenon data of the W field are marked as W.t The corresponding DB field ground temperature data is marked as T t ;

[0118] Step P403-3: Create condensation weather data W t , corresponding to ground temperature T t These two arrays; get formula (6):

[0119] f(W)=W t , T t >0 (6);

[0120] Step P403-4: Using W t and T t Check the authenticity of suspicious data. t >0, and the weather data W t Data recorded as “02” frost, “03” ice and “16” snow accumulation are suspected to be erroneous data.

[0121] like Fig.11 The comparison chart of ice and frost before and after quality control shows that there are many abnormal data before quality control. In the current business regulations, manual observation of ice and frost has been cancelled, but some stations in Xinjiang are still recording manually. This may be related to the fact that meteorological station personnel are not familiar with the business regulations for ice and frost recording. Business personnel can be organized to strengthen the study of new business regulations.

[0122] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the claims of this patent application.

Claims

1. A method for multi-factor coordinated meteorological data quality control of frozen soil, snow depth, and condensation weather phenomena in Xinjiang, characterized in that: The steps include: Step P1: Obtain full-factor meteorological data and monthly observation data of daily frozen ground, snow depth, and condensation weather phenomena from national ground-based meteorological observation stations in Xinjiang; Step P2: construct the relationship expressions between frozen soil, snow depth, condensation-type weather phenomena and meteorological elements respectively, and obtain the theoretical frozen soil depth, theoretical snow depth and theoretical condensation-type weather phenomena; Step P3: Compare and analyze the actual frozen soil depth, actual snow depth and actual condensation-type weather phenomena in the monthly observation data in step P1 with the theoretical frozen soil depth, theoretical snow depth and theoretical condensation-type weather phenomena in step P2, filter out suspicious data, and determine the rationality of the frozen soil, snow depth and condensation-type weather phenomenon elements.

2. According to claim 1, a method for multi-factor coordinated meteorological data quality control of frozen soil, snow depth, and condensation weather phenomena in Xinjiang, characterized in that: In step P1, 105 national meteorological stations with complete observation elements were selected from all meteorological stations in Xinjiang. The standard was that the obtained observation data had a complete sequence. At the same time, the selection of observation stations could reflect the local terrain, landform and climate characteristics.

3. The method for controlling the quality of multi-factor coordinated meteorological data of frozen soil, snow depth, and condensation weather phenomena in Xinjiang according to claim 2 is characterized in that: Step P1 includes the following: Step P101: Obtain full-factor meteorological observation monthly data from 105 national ground stations across Xinjiang; Step P102: Find the contents of the A, Z, W, and DB fields in the full-factor monthly meteorological observation data of 105 national ground stations across Xinjiang, and obtain daily observation data on frozen ground, snow depth, and condensation weather phenomena.

4. The method for controlling the quality of multi-factor coordinated meteorological data of frozen soil, snow depth, and condensation weather phenomena in Xinjiang according to claim 3 is characterized in that: In step P2, based on the significance analysis of frozen soil and weather phenomena and meteorological elements, it was found that there was a significant correlation between frozen soil and the corresponding ground temperature of 0 cm, 5 cm, and 15 cm, thus constructing the theoretical frozen soil depth and ground temperature correlation expression: f(T t ) = aT DB and a < 0, T DB ≤ 0(1); Among them, f(T t ) is the theoretical frozen soil depth, T DB is ground temperature, a is f(T t ) and T DB The correlation coefficient is negative.

5. The method for controlling the quality of multi-factor coordinated meteorological data of frozen soil, snow depth, and condensation weather phenomena in Xinjiang according to claim 3 is characterized in that: In step P2, based on the significance analysis of snow depth and weather phenomena and meteorological elements, it is found that there is a significant correlation between snow depth and snowfall weather phenomena, existing snow depth, and ground temperature, thereby constructing a correlation expression between theoretical snow depth and snowfall weather phenomena, existing snow depth, and ground temperature: f(z t )≥0, and: When T DB ≥0, then f(z t )=cW+Z+bT DB , c>0, W≥0, Z≥0, b<0(2); When T DB When <0, then f(z t )=cW+Z, c>0, W≥0, Z≥0(3); Among them, f(z t ) is the theoretical snow depth; W is the weather phenomenon of snow, and W ≥ 0; T DB is ground temperature; Z is the existing snow depth, and Z>0; b is f(z t ) and T DB The correlation coefficient of , and b<0, that is, f(z t ) and T DB Negatively correlated; c is f(z t ) and W, and c>0, that is, f(z t ) is positively correlated with c.

6. The method for controlling the quality of multi-factor coordinated meteorological data of frozen soil, snow depth, and condensation weather phenomena in Xinjiang according to claim 3 is characterized in that: In step P2, condensation weather phenomena include frost, ice and snow weather phenomena. Based on the significance analysis of condensation weather and meteorological elements of weather phenomena, it is found that there is a significant correlation between condensation weather and ground temperature. DB Negatively correlated, that is, when T DB When <0, condensation-like weather phenomena will theoretically occur.

7. The method for controlling the quality of multi-factor coordinated meteorological data of frozen soil, snow depth, and condensation weather phenomena in Xinjiang according to claim 1 is characterized in that: In step P3, it includes: Step P301: Perform data quality control on all-element meteorological data from 105 national ground stations, so that each meteorological element passes internal consistency check, time consistency check, climate limit value and spatial consistency check; Step P302: Obtain the actual permafrost depth, actual snow depth and actual condensation-like weather phenomena in the monthly observation data of Xinjiang on that day, compare them with the theoretical permafrost depth, theoretical snow depth and theoretical condensation-like weather phenomena of Xinjiang on that day, mark the suspected erroneous data, use the relevant data of the previous day and the next day to screen and correct the suspected erroneous data, and verify the authenticity of the data.

8. The method for controlling the quality of multi-factor coordinated meteorological data of frozen soil, snow depth, and condensation weather phenomena in Xinjiang according to claim 7 is characterized in that: For frozen soil data, Step P401-1: Filter out the frozen soil data in the A field and the ground temperature data in the DB field in the monthly data file, and use the frozen soil data in the A field to obtain the actual frozen soil depth data A. t At the same time, using the ground temperature data in the DB field, the theoretical frozen soil depth f(T t ); Step P401-2: The actual frozen soil depth data A t and theoretical frozen soil depth f(T t )Compare the data of these two arrays and traverse A t (t=1,2,···,n) frozen soil observation series data and theoretical frozen soil depth f(T t )(t=1,2,···,n) t ), we get formula (4): f(A t )=A t -f(T t ) (4); Step P401-3: Screen out f(A t ) is a negative value, and the actual frozen soil depth data A t Zero or missing, ground temperature 5cm T DB Data <0 are marked as suspicious frozen soil data; Step P401-4: Compare the selected suspicious frozen soil data with the difference f(A t-1 ) and the difference f(A t+1 ), check the suspicious data f(A t ) is true, if [f(A t )< <f(A t-1 )]∪[f(A t )< <f(A t+1 )] value, it is suspected erroneous data, otherwise it is correct data.

9. The method for controlling the quality of multi-factor coordinated meteorological data of frozen soil, snow depth, and condensation weather phenomena in Xinjiang according to claim 7 is characterized in that: For snow depth data, Step P402-1: Filter out the existing snow depth data in the Z field, the snowy weather phenomenon data in the W field, and the ground temperature data in the DB field in the monthly data file; Step P402-2: Establish actual snow depth data Z t and theoretical snow depth f(z t ) These two arrays: the actual snow depth data Z is obtained from the existing snow depth data in the Z field and the snowy weather phenomenon data in the W field t ; Based on the ground temperature data in the DB field, the theoretical snow depth f(z t ) data; at the same time, the actual snow depth data Z t and theoretical snow depth f(z t ) to process the data, which is divided into two cases: snow depth and no snow depth. Snow depth is represented as 1, and no snow depth is represented as 0; Step P402-3: Compare the data of the two arrays and traverse Z t (t=1,2,···,n) Actual snow depth observation sequence data and f(z t )(t=1,2,···,n) theoretical snow depth data, and obtain the difference between the two data sets f(Z t ), we get formula (5): f(Z t )=Z t -f(z t ) (5); Step P402-3: If the snow depth difference f(Z t ) data is 1 or -1, it is marked as suspicious snow depth data; Step P402-4: For the suspicious snow depth data that has been screened out, use the ground temperature data and snow depth data Z of the previous day to t-1 and the ground temperature data and snow depth data of the next day Z t+1 Screen the snow depth and monthly meteorological observation data to verify the authenticity of the suspected erroneous data: When the marked ground temperature is greater than zero degrees, there is suddenly a deep snow T t-1 >0,Z t-1 =0)∩(T t >0,Z t =1)∩(T t+1 >0,Z t+1 =0), the ground temperature is below zero and there is no snow depth (T t-1 <0,Z t-1 =1)∩(T t <0,Z t =0)∩(T t+1 <0,Z t+1 =1), and combined with the weather phenomenon of the day, the "70" field marks that there is no snow and suddenly there is snow depth W t =0,Z t =1. When it snows suddenly, the snow depth is not deep. t >0,Z t =0, it is suspected erroneous data, otherwise it is correct data.

10. The method for controlling the quality of multi-factor coordinated meteorological data of frozen soil, snow depth, and condensation weather phenomena in Xinjiang according to claim 7, characterized in that: For condensation weather phenomena: Step P403-1: Filter out the weather phenomenon data in the W field and the ground temperature data in the DB field in the monthly data file; Step P403-2: Data processing: the fields with "02" frost, "03" ice and "16" snow in the weather phenomenon data of the W field are marked as W. t The corresponding DB field ground temperature data is marked as T t ; Step P403-3: Create condensation weather data W t , corresponding to ground temperature T t These two arrays; get formula (6): f(W)=W t ,T t >0 (6); Step P403-4: Using W t and T t Check the authenticity of suspicious data. t >0, and the weather data W t Data recorded as "02" frost, "03" ice and "16" snow accumulation are suspected to be erroneous data.

Images