CN117370714B - Representative station quantitative determination method - Google Patents

Abstract

The invention discloses a method for quantitatively determining representative stations, which uses the correlation coefficient and the representative area as indicators to determine the optimal representative station through quantitative optimization and identification. This representative station determination method has the advantages of being quantitative, objective, and highly operable, and can be used for resource evaluation and scientific research in the fields of meteorology, environment, hydrology, ecology, etc.

Description

A quantitative determination method for representative stations

Technical field

The invention belongs to the technical field of meteorological environment (environmental protection), and in particular relates to a method for determining a meteorological environment ecological observation representative station. It can be used for resource evaluation and scientific research in the fields of meteorology, environment, ecology, hydrology and other fields.

Background technique

Meteorological, environmental, ecological, hydrological and other departments conduct continuous monitoring and data collection of relevant elements through observation stations. For example, weather stations use various measurement equipment to observe and collect meteorological elements such as temperature, humidity, air pressure, wind speed, wind direction, and rainfall in the atmospheric environment. These data are of great significance for weather forecasting, climate resource evaluation, meteorological disaster prevention and reduction, etc. important supporting role. With the improvement of my country's comprehensive national strength, meteorological, environmental, ecological, hydrological and other departments have continued to increase their efforts in the construction of observation stations, and the density of various observation stations has gradually increased. In practical applications and scientific analysis, it is often necessary to select appropriate ones. A representative station is an observation station that can effectively represent the overall situation of a certain observation element within a certain range. Scientific and reasonable selection of representative stations is of great significance in getting twice the result with half the effort. Traditional representative station selection lacks quantitative indicators and has certain limitations in practical application.

Contents of the invention

Purpose of the invention: In view of the problems existing in the existing technology, the present invention designs a method for quantitative determination of representative stations. Through quantitative optimization and identification, the optimal representative station is determined. The method of the present invention can be used in the fields of meteorology, environment, ecology, hydrology and other fields. Resource evaluation and scientific research, etc.

Technical solution: In order to achieve the above-mentioned object of the invention, the present invention adopts the following technical solution: a representative station quantitative determination method, including the following steps:

Step S1, select target station

Among all stations in the study area, a target station _Pi is randomly selected;

Step S2: Determine the correlation between the target station and surrounding stations

(1) With the target station _Pi as the center, select the M stations closest to the target station _Pi as the surrounding stations _{P ij} of the target station Pi (j=1...M);

(2) Calculate the correlation coefficient of _{the observation element sequence of the target station Pi and its surrounding station P ij ,} and obtain the square sequence of the correlation coefficient r _ij ² (j=1...M);

Step S3: Determine the effective surrounding stations of the target station

Extract all values r _ik ² (k=1...mi) that are greater than or equal to the given threshold T in the square sequence of correlation coefficients r _ij ² (j=1... _M ), and their corresponding surrounding stations P _ik (k=1... m _i ) as an effective surrounding station of target station _Pi ;

Step S4: Calculate the representative factor of the target station

Calculate the representative factor R _i (coefficient of determination factor) and G _i (representative area factor) of the target station P _i . The calculation formula is:

(1);

(2);

In the formula, R _i is the determination coefficient factor; G _i is the representative area factor; r _ik is the correlation coefficient of the observation element sequence of the target station P _i and its effective surrounding station P _ik , m _i is the effective surrounding station of the target station P _i Number of stations; S _i is the Thiessen polygon area corresponding to the target station _Pi , and S _ik is the Thiessen polygon area corresponding to the effective surrounding station P _ik .

Step S5, select a new target station, repeat steps S2-S4, traverse all stations, and obtain the representative factors of all stations in the study area;

Step S6: Determine the representativeness index of each station

(1) Normalization processing of the representative factors of each station. Normalize the representative factors R _i (coefficient of determination factor) and G _i (representative area factor) of the target station _Pi , and obtain the normalized values respectively. Values A _i and B _i ;

(2) The representativeness index F _i of each station is calculated by the following formula (3),

(3)

In the formula, F _i is the representative index of the target station _Pi , W _A and W _B are the weights of the representative factors respectively, and N is the total number of stations in the study area;

All stations were traversed to determine the representativeness index of each station in the study area.

Step S7: Determine the number of representative votes for each station

Compare the representative index F _ik of the target station P _i with its effective surrounding station P _ik (k=1...mi ₎ , and statistically The number of occurrences C _i is used as the representative number of votes for the target station _Pi . Traverse all stations and determine the number of representative votes for each station;

Step S8: Determine the representative station

Set the representative vote threshold D, and compare _{the representative vote number C i of} the target station Pi with D. If C _i ≥ D, the target station _Pi is the representative station. Traverse all stations and determine whether each station is a representative station.

Further, the station is a meteorological observation station, an environmental observation station or a hydrological observation station.

Further, in step S2, the value of the number M of search stations closest to the target station _Pi is set to 20~40; the square correlation coefficient threshold T is set to 0.6~0.9.

Further, the observation element sequence of the target station Pi and its surrounding station _{P ij in} step S2 is the observation element sequence of a meteorological observation station, an environmental observation station or a hydrological observation station, such as: the temperature observation sequence of a meteorological station, the observation sequence of an environmental observation station PM2.5 observation sequence.

Further, the correlation coefficient of the observation element sequence of the target station P _i and its surrounding station P _ij in step S2 is calculated by the following formula:

(4);

In the formula, x _l (l=1...n) is the observation element sequence of target station _Pi , is the average value of x _l (l=1...n); y _l (l=1...n) is the observation element sequence of P _i 's surrounding station P _ij ,/> is the average value of y _l (l=1...n); n is the length of the observation element sequence, that is, the number of samples.

Further, the representative factors of each station described in step S6 are normalized according to the following formula,

(5);

(6);

In the formula, R _min and R _max are respectively the minimum and maximum values of the determination coefficient factor R _i (i=1...N), N is the total number of stations in the study area; G _min and G _max are respectively the representative area factor G _i ( i=1…N) minimum and maximum values.

Further, the weights W _A and W _B of the representative factors described in step S6 have a numerical range of 0.1~0.9, and should satisfy W _A + W _B =1.0. When the two factors have equal weights, W _A and W _B simultaneously takes the value 0.5.

Beneficial effects: Compared with the existing technology, the present invention proposes a new scientific and reasonable quantitative determination method for representative stations, which is of great significance to resource evaluation and scientific research. The representative station determination method provided by the present invention has the advantages of quantitative, objective, strong operability, easy promotion, etc., and can be widely used in meteorology, hydrology, environment, ecological and environmental protection and other fields.

Description of the drawings

Figure 1 is a schematic flow chart of the representative station quantitative determination method according to the present invention;

Figure 2 is a schematic diagram of the Tyson polygon of the weather station in the research area according to the embodiment of the present invention;

Figure 3 is a Thiessen polygon schematic diagram of the target station and its effective surrounding stations according to the embodiment of the present invention;

Figure 4 is a representative station of rainfall in the study area according to the embodiment of the present invention.

Detailed ways

The present invention will be further clarified below in conjunction with the accompanying drawings and specific examples. It should be understood that these examples are only used to illustrate the present invention and are not intended to limit the scope of the present invention. After reading the present invention, those skilled in the art will be familiar with various aspects of the present invention. Modifications in the form of equivalents fall within the scope defined by the appended claims of this application.

As shown in Figure 1, a schematic flow chart of the quantitative determination method of representative stations according to the present invention, the method steps are as follows:

1. Select target site

Among all stations in the study area, a target station _Pi is randomly selected.

2. Calculate the correlation between the target station and surrounding stations

① With the target station _Pi as the center, calculate the distances between other stations and the target station, and select the M stations closest to the target station _Pi as the surrounding _stations P _ij of the target station Pi (j=1...M).

② Calculate the correlation coefficient r _ij of the observation element sequence of the target station P _i and its surrounding stations P _ij (j=1...M), and calculate it according to the correlation coefficient formula:

(1);

In the formula, x _l (l=1...n) is the observation element sequence of target station _Pi , is the average value of x _l (l=1...n); y _l (l=1...n) is the observation element sequence of P _i 's surrounding station P _ij ,/> is the average value of y _l (l=1...n); n is the length of the observation element sequence, that is, the number of samples. The observation element sequence is the observation element sequence of a meteorological observation station, an environmental observation station or a hydrological observation station, such as: the temperature observation sequence of a meteorological station, and the PM2.5 observation sequence of an environmental observation station.

On this basis, the correlation coefficient square sequence r _ij ² (j=1…M) is obtained.

3. Determine the valid surrounding stations of the target station

Extract all values r _ik ² (k=1...mi) in r _ij ² (j=1...M ₎ that are greater than or equal to the given threshold T and their corresponding surrounding stations P _ik (k=1...m _i ), and then P _ik is called the "effective surrounding station" of the target station _Pi .

4. Calculate the representative factor of the target station

Calculate the representative factor R _i (coefficient of determination factor) and G _i (representative area factor) of the target station P _i . The calculation formula is:

(2);

(3);

In the formula, R _i is the determination coefficient factor; G _i is the representative area factor; r _ik is the correlation coefficient of the observation element sequence of the target station P _i and its effective surrounding station P _ik , m _i is the effective surrounding station of the target station P _i Number of stations; S _i is the Thiessen polygon area corresponding to the target station _Pi , and S _ik is the Thiessen polygon area corresponding to the effective surrounding station P _ik . The Thiessen polygon schematic diagram of the study area is shown in Figure 2, and the Thiessen polygon schematic diagram of the target station and its effective surrounding stations is shown in Figure 3.

5. Select new target site

Select a new target station, repeat steps 2-4, traverse all stations, and obtain the representative factors of all stations in the study area.

6. Calculate the representativeness index of each station

①The representative factors of each station are normalized.

The representative factors R _i (coefficient of determination factor) and G _i (representative area factor) of the target station _Pi are normalized according to the following formula:

(4);

(5);

In the formula, A _i is the normalized value of the determination coefficient factor R _i , B _i is the normalized value representing the face factor G _i , R _min and R _max are respectively the determination coefficient factor R _i (i=1...N) The minimum and maximum values of , N is the total number of sites in the study area. G _min and G _max represent the minimum and maximum values of the face factor G _i (i=1...N) respectively.

② Calculation of representativeness index of each station

The representativeness index F _i of each station is calculated by the following formula (6),

(6);

F _i is the representative index of the target station P _i , W _A and W _B are the weights of the representative factors, their values are 0.1~0.9, and should satisfy W _A + W _B =1.0, when the two factors have equal weights , W _A and W _B take the value 0.5 at the same time. All stations were traversed to determine the representativeness index of each station in the study area.

7. Determine the number of representative votes for each station

For the target station P _i , compare its representative index F _i with the representative index F _ik (k=1...mi ₎ of its effective surrounding stations P _ik (k=1...mi ₎ , and statistically The number of occurrences C _i , C _i is the representative number of votes for the target station _Pi . All stations were traversed to determine the representative number of votes for all stations in the study area.

8. Determine representative station

Set the representative vote threshold D. For the target station _Pi , compare its representative vote C _i with D. If , then _Pi is the representative station, and vice versa.

By traversing all stations, all representative stations in the study area can be selected. By adjusting the size of the representative vote threshold D, the density of representative stations can be adjusted.

Implementation case:

Taking 200 weather stations in Nanjing City as an example, the implementation process of the present invention is explained. According to the steps of the present invention, the rainfall representative station in Nanjing is determined. Set the value of the number of search stations M to 20, the square correlation coefficient threshold T to 0.85, and the representative factor weights W _A and W _B to take the value 0.5 at the same time. Use the monthly rainfall of 200 weather stations in Nanjing from 2018 to 2022. Observe the sequence, calculate the correlation coefficient between the target station and its surrounding stations, use ArcGIS to generate the Tyson polygon of each weather station, and then obtain the corresponding Tyson polygon area of each station. According to the implementation steps of the present invention, the determined rainfall representative station in the study area is shown in Figure 4 .

Taking the above-mentioned ideal embodiments of the present invention as inspiration and through the above description, relevant workers can make various changes and modifications without departing from the scope of the technical idea of the present invention. The technical scope of the present invention is not limited to the content in the description, and must be determined based on the scope of the claims.

Claims (6)

1. A representative station quantitative determination method, characterized by comprising the steps of:

step S1, selecting a target station

The stations are weather observation stations, environment observation stations or hydrologic observation stations, and one target station P is randomly selected from all stations in a research area _i ；

Step S2, determining the relativity of the target station and the peripheral station

(1) By destination station P _i For the center, select and target station P _i M stations closest to the station as destination stations P _i Peripheral station P of (a) _ij （j=1…M）；

(2) Calculating the destination P _i With peripheral station P _ij The correlation coefficient of the observation element sequence of (2) to obtain a square sequence r of the correlation coefficient _ij ² （j=1…M）；

Step S3, determining the effective surrounding station of the target station

Extracting square sequence r of correlation coefficient _ij ² All values r in (j= … M) equal to or greater than a given threshold T _ik ² （k=1…m _i ) Its corresponding peripheral station P _ik （k=1…m _i ) As destination station P _i Is effective for the peripheral station;

step S4, calculating the representative factor of the target station

By destination station P _i Square sequence r of correlation coefficients with peripheral stations _ij ² Average value R of (2) _i And destination station P _i Is the representative area factor G of (2) _i As a destination station P _i Is calculated as:

（1）

（2）

wherein R is _i Determining coefficient factors; g _i Is a representative area factor; r is (r) _ik For destination station P _i And an effective peripheral station P _ik Correlation coefficient of observation element sequence, m _i For destination station P _i The number of active peripheral stations; s is S _i For destination station P _i Corresponding Thiessen polygonal area, S _ik For effective peripheral station P _ik Corresponding Thiessen polygonal area;

step S5, selecting a new target station, repeating the steps S2-S4, traversing all stations, and obtaining representative factors of all stations in a research area;

step S6, determining representative index of each station

(1) Normalization processing of representative factors of each station, and target station P _i Representative factor R of (2) _i (determining coefficient factor) and G _i Normalized (representing area factor) to obtain normalized value A _i And B _i ；

(2) Representative index F of each station _i Is calculated by the formula (3),

（3）

wherein F is _i For destination station P _i Representative index of W _A 、W _B Weights of representative factors respectively, wherein N is the total number of sites in a research area;

traversing all stations, and determining a representative index of each station in the study area;

step S7, determining representative ticket numbers of each station

Will target station P _i And an effective peripheral station P _ik （k=1…m _i ) Representative index F of (2) _ik Comparing and countingNumber of occurrences C _i As a destination station P _i Representative ticket numbers of (a); traversing all stations, and determining a representative ticket number of each station in a study area;

step S8, determining representative station

Setting a representative ticket number threshold D to enable the target station P to be _i Representative ticket number C _i Comparing with D, if C _i Not less than D, the destination station P _i Namely a representative station; all stations are traversed and a determination is made as to whether each station of the study area is a representative station.

2. The representative station quantitative determination method according to claim 1, wherein: step S2 the destination station P _i Setting the value of the number M of the nearest searching stations to be 20-40; the correlation coefficient square threshold T is set to 0.6-0.9.

3. The representative station quantitative determination method according to claim 1, wherein: step S2 the destination station P _i With peripheral station P _ij The observation element sequence of (a) is an observation element sequence of a weather observation station, an environment observation station or a hydrological observation station.

4. The representative station quantitative determination method according to claim 1, wherein: step S2 the destination station P _i With peripheral station P _ij Is a part of the observation element of (a)The correlation coefficient of the sequence is calculated as:

（4）

wherein x is _l (l= … n) is the destination station P _i Is characterized in that the sequence of the observation elements of (a),is x _l An average value of (l= … n); y is _l (l= … n) is P _i Peripheral station P of (a) _ij Is a sequence of observation elements,/->Is y _l An average value of (l= … n); n is the length of the sequence of observation elements, i.e. the number of samples.

5. The representative station quantitative determination method according to claim 1, wherein: the representative factors of each station in step S6 are normalized as follows,

（5）

（6）

wherein R is _min 、R _max Respectively determining coefficient factors R _i (i= … N) minimum and maximum values, N being the total number of sites in the investigation region; g _min 、G _max Respectively representing area factors G _i Minimum and maximum values of (i= … N).

6. The representative station quantitative determination method according to claim 1, wherein: step S6 weight W of the representative factor _A 、W _B The numerical range is 0.1-0.9, and W should be satisfied _A +W _B =1.0, when two factorsWhen the weights are equal, W _A 、W _B And simultaneously takes a value of 0.5.