JP2007248355A - Device and method for estimating weather data - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a device for uniquely estimating weather elements at an arbitrary three-dimensional grid point by employing only weather observation data contained, in an object range, without adding virtual weather observation data, independently of a selection method; and a method therefor. <P>SOLUTION: In a weather data estimating and calculating section 12, a micro weighing factor judging section 122 judges whether weighing factors of all observation points for each grid point are a micro threshold value previously specified or less and calculates the ratio of the total number of micro weighing factor grid points* to that of all grid points; a micro weighing factor correction process section 123 corrects weighing factors among that of the micro weighing factor grid points in accordance with the previously specified rule so that a weighing factor of at least one observation point is larger than the micro weighing factor; and thus, a weather data primary estimation processing section 124 can obtain proper a primary estimation value from weather data at all grid points. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a meteorological data estimation technique used for local meteorological surveys, atmospheric diffusion predictions, and the like, and in particular, an apparatus and a method capable of uniquely estimating meteorological data at an arbitrary three-dimensional lattice point from only limited meteorological observation data About.

  Meteorological elements (wind direction, wind speed, rainfall, etc.) observation data at AMeDAS and meteorological observatory for local meteorological surveys and atmospheric diffusion predictions taking into account the effects of complex topography and spatial and temporal changes in weather conditions Numerous numerical simulations using the are performed. In such a simulation, it is necessary to divide the target three-dimensional space area into a grid and accurately estimate the weather element at each grid point. An objective analysis method is often used as a method for estimating the meteorological element of a three-dimensional lattice point with relatively high accuracy without depending on a method with a large calculation load. In the objective analysis method, as a first step, a primary estimated value is obtained by averaging meteorological observation data included in a target region on a three-dimensional grid point with a weight taking into account the influence of distance and topography.

  Next, as a second step, the meteorological data primary estimated value at each grid point is corrected by objective analysis according to the laws of physics. For example, as an objective analysis method of three-dimensional wind velocity field for atmospheric dispersion simulation in complex terrain, using the mass conservation law as a constraint, the primary estimate of the wind velocity field at each grid point and the total amount of correction are minimized Thus, a so-called mascon model that is calculated by a variational method is widely applied. A technique for improving the accuracy of general objective analysis includes a technique disclosed in Patent Document 1.

  On the other hand, in order to improve the estimation accuracy of the weather data of the three-dimensional lattice points, it is important to increase the accuracy of the primary estimation value. For example, as a method for accurately obtaining first-order wind field estimates for objective analysis of 3D wind field using the above-mentioned mascon model from observation data of limited wind speeds and directions, ground observation points that exist on complex terrain Non-Patent Document 1 discloses a method of weighting and averaging the above data on a three-dimensional lattice point using a product of three kinds of weight coefficient components as a function of horizontal distance, vertical distance, and terrain barrier.

Japanese Patent Application Laid-Open No. 2005-249662 Journal of Nuclear Science and Technology, Vol.25, No.9, pp.721〜730 (1988).

  When terrain is complicated and relatively narrow (about 50km square), and meteorological observation data is used to estimate meteorological data on a three-dimensional grid using the objective analysis method described above, meteorological observation points included in the target range The weighting factors of observation data set from each meteorological observation point at some grid points are all close to zero due to the fact that there are few, their positions are not evenly distributed, and there is no wind direction data over the sky. Since it becomes a minute value, a weighted average cannot be calculated numerically, and a primary estimated value cannot be obtained, so objective analysis is often impossible. As countermeasures, virtual weather data can only be artificially added by trial and error, making it difficult for beginners to perform objective analysis of weather data. In addition, even if virtual weather data is added to obtain a result, there is a problem that the obtained result is not unique depending on the position and numerical value of the added virtual weather data.

  The object of the present invention is to perform simulation for atmospheric dispersion prediction and local weather survey using meteorological observation data, without adding virtual meteorological data, regardless of how to select the target range. An object of the present invention is to provide an apparatus and method capable of uniquely estimating meteorological data at an arbitrary three-dimensional lattice point using only meteorological observation data included in a range.

  In order to solve the above-mentioned problems, the present invention estimates meteorological element data at each grid point obtained by dividing the target space area into a grid pattern from meteorological element observation data at a plurality of points included in the target space and time domain. A meteorological data estimation device for calculating a weighting coefficient of meteorological element observation data at each point with respect to a grid point, and a primary estimated value of the meteorological element at each grid point for meteorological element observation at each point Meteorological data primary estimation processing means for calculating the data by multiplying the data by the weighting factor, and calculating the meteorological element data at each grid point by correcting the result based on the physical law that the meteorological element data follows In the meteorological data estimation calculation means composed of the meteorological data objective analysis processing means, for each grid point, the weighting factor of the observation data set from each meteorological observation point It is determined whether or not all of them are below a preset minute threshold, and one or more meteorological observation points are obtained so that weather data estimates can be obtained for such lattice points (weight coefficient minute lattice points). There is provided a meteorological data estimation device comprising: a minute weight coefficient correction processing means for correcting a weight coefficient according to a predetermined rule so that a set weight coefficient exceeds the minute threshold value.

  By providing the minute weight coefficient judging means, all the weight coefficients set from each meteorological observation point at some grid points (weight coefficient minute grid points) are less than the preset minute threshold value and numerically weighted. If the weighted average cannot be calculated, the weighting factor is determined in advance by the weighting factor correction processing means so that the weighting factor of one or more observation points exceeds the threshold value for the weighting factor minute grid point. Therefore, it is possible to calculate the weighted average of the observation data even with the weight coefficient micro lattice points using only the meteorological observation data included in the target range without adding virtual weather data. Thus, even a beginner can easily obtain a primary estimated value of weather data.

  Specifically, the present invention relates to meteorological element data that estimates meteorological element data at each grid point obtained by dividing the target space area into a grid pattern from meteorological element observation data at a plurality of meteorological observation points included in the target space and time domain. In the data estimation device, the weighting factor of meteorological element data at each point corresponding to each grid point is at least one of the horizontal distance, vertical height, and topography barrier height between each grid point and each weather observation point, or An observation data weighting coefficient calculation means for calculating the product of these components by assigning a weighted coefficient (weighting coefficient component) closer to each grid point for these multiple points, and an estimate of meteorological element data at each grid point Meteorological data primary estimation processing means that calculates the average of the meteorological element observation data of each meteorological observation point by the above-mentioned calculated observation data weighting coefficient, and each grid point. Minute weight coefficient judging means for judging whether the maximum value of the observation data weighting coefficient at the observation point is less than or equal to a predetermined threshold value, and the calculated maximum value of the observation data weighting coefficient is less than or equal to the minute threshold value If it is determined that the weight coefficient of each weather observation point is used by using the product of only the observation data weight coefficient components whose maximum value exceeds the minute threshold, or by assigning a constant without weighting Minute weight coefficient correction processing means for correcting and setting so that the maximum value exceeds the minute threshold value, and corrected weather for correcting the weather data estimated by the weather data primary estimation process using the corrected weight coefficient There is provided a meteorological data estimation device characterized by comprising data estimation processing means.

The present invention also provides meteorological data estimation for estimating meteorological element data at each grid point obtained by dividing the target space area into a grid from meteorological element observation data at a plurality of meteorological observation points included in the target space and time domain. In the meteorological data estimation program used in the equipment, the weighting factor of meteorological element observation data at each point corresponding to each grid point is set to at least the horizontal distance, vertical height, and topography barrier height between each grid point and each meteorological observation point. Any one of these or a plurality of these, a weighted coefficient component (weighting factor component) is assigned to each lattice point, and an observation data weighting factor calculation program that calculates the product of those components, Meteorological data calculated by averaging the estimated values of the meteorological element data by multiplying the meteorological element observation data at each meteorological observation point by the observed data weighting factor calculated above. A next estimation processing program, a minute weighting factor judgment program for judging whether the maximum value of the aforementioned observation data weighting factor at each meteorological observation point is less than or equal to a preset minute threshold value, and the calculated observation When it is determined that the maximum value of the data weighting factor is less than or equal to the minute threshold value, use the product of only the observation data weighting factor components whose maximum value exceeds the minute threshold value, or weight all observation points. Without assigning a constant, a fine weight coefficient correction processing program that corrects and sets the weight coefficient maximum value of each weather observation point to exceed the fine threshold value, and the weight coefficient that is corrected and set are used. A weather data estimation processing program for correcting weather data estimated by the weather data primary estimation processing. Providing weather data estimation program for use in data estimation device.

The present invention also provides meteorological data estimation for estimating meteorological element data at each grid point obtained by dividing the target space area into a grid from meteorological element observation data at a plurality of meteorological observation points included in the target space and time domain. In the meteorological data estimation method by the device, the observation data weight coefficient calculation means calculates the weight factor of the meteorological element observation data at each point corresponding to each grid point, at least the horizontal distance between each grid point and each weather observation point, the vertical height A coefficient (weighting coefficient component) weighted as the grid points are closer to each grid point for either of the height and the height of the barrier, is calculated as the product of those components, and the weather data primary estimation processing means The mean value of the meteorological element data estimated at each grid point is multiplied by the above-mentioned calculated observation data weighting factor multiplied by the meteorological element observation data at each meteorological observation point. And determine whether the maximum value of the aforementioned observation data weighting factor at each meteorological observation point is less than or equal to a preset microthreshold value for each grid point by means of the microweighting factor determination means, and fine weighting factor correction processing When it is determined by the means that the maximum value of the calculated observation data weighting coefficient is less than or equal to the minute threshold value, the product of only the observation data weighting coefficient component whose maximum value exceeds the minute threshold value is used, Alternatively, the correction value is set so that the maximum value of the weighting coefficient of each weather observation point exceeds the minute threshold by assigning a constant without weighting all the observation points, and the correction setting is performed by the correction weather data estimation processing means. The weather data estimated by the weather data primary estimation process is corrected using the weighted weighting factor. To provide a data estimation method.

  According to the present invention, when performing simulation for atmospheric dispersion prediction or local weather survey using meteorological observation data, only the meteorological observation data included in the target range is used, regardless of how to select the target range. Because it is possible to uniquely estimate meteorological data at any three-dimensional grid point without adding virtual meteorological observation data by trial and error, there is an effect that even beginners can easily and efficiently perform objective analysis of meteorological data. .

  In a preferred embodiment of the present invention, in the meteorological data estimation device, when the minute weight coefficient determination means finds that one or more weight coefficient micro lattice points exist, the minute weight coefficient determination means calculates the weight data. Weight coefficient micro grid point information display means for displaying or outputting weight coefficient micro grid point information such as the total number of the weight coefficient micro grid points and the ratio to the total grid points on an output device such as a screen; Based on grid point information, user instruction input means that can select whether to perform the minute weighting coefficient correction process, or to add meteorological observation data of a new point or an estimated value thereof, and a weather observation of the new point by the user Interactive user interface means comprising meteorological observation data addition input means when it is selected to add data or its estimated value Providing weather data estimation device characterized by comprising.

  Thereby, when the weighting coefficient microgrid points exist, the user performs the microweighting coefficient correction processing based on the weighting coefficient microgrid point information, or the meteorological observation data of the new point or the estimated value thereof. Since it is possible to select whether to add, it is possible for an expert to obtain a more accurate estimate of weather data.

  Hereinafter, a weather data estimation program, a method including processing steps realized by the weather estimation device, and a weather data estimation program for causing a computer to execute the processing steps will be described with reference to the drawings.

FIG. 1 shows an embodiment of the configuration of a meteorological data estimation apparatus according to the present invention.
The meteorological data estimation device 100 includes an input device 1 for inputting data, an output device 2 such as a display and a printer, a storage device 3 including various programs such as a hard disk and a CD-ROM, and an arithmetic processing device 10. . The arithmetic processing device 10 includes a data setting unit 11, a weather data estimation calculation unit 12, an interactive user interface unit 13, and a weather data estimation result display unit 14.

  The data setting unit 11 reads and sets data necessary for objective analysis of weather data from the input device 1 or the storage device 3. The map data reading unit 111 reads raster or vector map data from the storage device 3 and displays it on the screen as a two-dimensional map. The three-dimensional space includes the horizontal distance, the vertical height and the terrain barrier specified by the user on the screen. The weather estimation range setting unit 112 reads and sets the weather estimation range and the time range of the date and time. Next, according to, for example, the number of grids in the vertical, horizontal, and height directions designated by the user, the calculation grid generation unit 113 sets three-dimensional calculation grid points. The altitude data of the selected three-dimensional weather estimation range thus selected is read by the altitude data reading unit 114, and the meteorological observation data and the meteorological observation position included in the space and time range are read from the storage device 3 by the meteorological observation data input unit 115, respectively. Read.

  Estimate and calculate meteorological elements (wind direction, wind speed, temperature, rainfall, etc.) on the three-dimensional grid points set by the data setting unit 11 by adding altitude data to the meteorological observation data read by the data setting unit 11 The meteorological data estimation calculation unit 12 includes an observation data weighting factor calculation unit 121, a minute weighting factor determination unit 122, a minute weighting factor correction processing unit 123, a weather data primary estimation processing unit 124, and a meteorological data objective analysis processing unit 125. Consists of a weather data estimation processing unit.

  In the observation data weight coefficient calculation unit 121, the weight factor of meteorological element observation data at each observation point with respect to each lattice point is set as the horizontal distance, vertical distance, and topography barrier height between each lattice point and each weather observation point (observation point). Calculated as a function such as In the following embodiment, the horizontal distance, the vertical distance, and the terrain barrier will be described as three factors. However, the weighting factor of the meteorological element observation data can be set for one or more of them, From the above accuracy, the above three factors are desirable.

  In the minute weighting factor determination unit 122 that is a feature of the embodiment, whether the maximum value of the weighting factor of each observation point calculated for each lattice point is equal to or less than a preset minute threshold value (ε). Is calculated, and the total number of such grid points (weight coefficient micro grid points) and the ratio to the total grid points are calculated. This result is used as a basis for determining whether or not the user performs a minute weight coefficient correction process described later in an interactive user interface unit described later.

  Furthermore, in the minute weight coefficient correction processing unit 123 which is the greatest feature of the embodiment, the weight coefficient fine grid points will be described in detail later, but roughly, the weight coefficient is corrected according to a predetermined rule, The weighting coefficient of one or more observation points is set so as to exceed the minute threshold value (ε). As a result, the meteorological data primary estimation processing unit 124 multiplies the weather element observation data at each observation point by the corrected weighting factor and averages the data on all grid points to obtain a reasonable primary estimation value of the meteorological data. The grid point weather data (primary estimation result) 34 is output to the storage device 3. The meteorological data objective analysis processing unit 125 corrects the primary estimation result obtained above based on the physical law according to the meteorological element data, thereby calculating the weather at each grid point (that is, the point corresponding to each grid point). The elements are calculated and output to the storage device 3 as lattice point meteorological data (objective analysis results) 35. This output result is useful as input data for local weather prediction and atmospheric dispersion simulation.

  As described above, in the meteorological data estimation apparatus according to the embodiment, the weighting factor is automatically corrected by the minute weighting factor correction processing unit 123 for the weighting factor minute grid point, so that an objective analysis result of the weather data is always obtained. . On the other hand, in the conventional meteorological data estimation device, since the weighting coefficient of all observation points is equal to or less than a minute threshold value (ε) at the weighting coefficient minute grid points, the meteorological data primary estimation processing unit 124 performs the meteorological data at each observation point. When the element observation data is multiplied by the weighting factor and averaged, both the denominator and numerator are small values close to zero, and the primary estimation value of the meteorological data cannot be calculated, and the objective analysis result cannot be obtained.

  The interactive user interface unit 13, which is another feature of the embodiment, includes the total number of weight coefficient micro lattice points calculated by the micro weight coefficient determination unit 122, the ratio to the total number of lattice points, the positions of the lattice points, and the like. The weight coefficient micro grid point information display unit 131 for displaying the information of the above, the user performs the micro weight coefficient correction processing based on the weight coefficient micro grid point information, or the meteorological observation data of a new point or its estimated value A user instruction input unit 132 that can select whether to add, or an observation data addition input unit 133 that performs input when the user selects to add meteorological observation data of the new point or its estimated value. On the other hand, when the user selects automatic correction of the minute weighting coefficient by the user instruction input unit 132, the weighting coefficient is automatically corrected by the minute weighting coefficient correction processing unit 123. In this way, the interactive user interface unit 13 allows the user to determine the validity of the objective analysis based on the weather observation data from the weighting coefficient micro-grid point information, so that the degree of freedom of the user for improving the accuracy of the objective analysis is increased. To increase. For example, if the weighting coefficient micro-grid points are 50% or more of the total grid points, it is possible to perform objective analysis by adding new meteorological observation points or estimating and adding virtual meteorological observation data by some appropriate method. Accuracy can be increased.

  The meteorological data estimation result display unit 14 outputs the grid point meteorological data (primary estimation result) 34 output to the storage device 3 by the meteorological data primary estimation processing unit 124 to the output device 2 such as a display or a printer. The display unit 141 and the lattice point meteorological data (objective analysis result) 35 output to the storage device 3 by the meteorological data objective analysis processing unit 125 are also composed of the objective analysis result display unit 142 that outputs the same to the output device 2. In these display units, for example, a map image, a wind speed vector diagram, an air temperature isotherm, and the like can be combined and displayed on a display.

  FIG. 2 shows an embodiment of a processing flow of the weather data estimation calculation unit 12 and the interactive user interface unit 13 which are features of the embodiment of the configuration example of the weather data estimation apparatus of FIG. First, the weight coefficient Wn (i, j, k) of the observation point n with respect to the lattice point (i, j, k) is calculated for all observation points (n = 1 to N), and their maximum value Wmax (i, j). , k) is calculated (step 201). Wn (i, j, k) is given as a function of the horizontal distance, vertical distance, topography barrier height, etc. between the lattice point (i, j, k) and the observation point n. A functional function has been proposed. Next, the minute weight coefficient determination unit 122 in FIG. 1 determines whether or not the maximum value Wmax (i, j, k) of the weight coefficient at each observation point is equal to or smaller than a predetermined small threshold value (ε). (Step 202), and when Wmax (i, j, k) ≦ ε, the weighting coefficient of all observation points is equal to or less than ε. Count (step 203). Here, the minute threshold value ε is appropriately set to a value that is much smaller than 1 and not less than 0. The larger this value is, the more the number of weight coefficient micro lattice points is. The above process is repeated for all grid points above the ground surface (step 204).

  Next, the minute weight coefficient determination unit 122 in FIG. 1 determines whether or not the total number of weight coefficient micro lattice points is larger than 0, that is, whether or not there are weight coefficient micro lattice points (step 205). The interactive user interface unit 13 and the minute weight coefficient correction processing unit 123 in FIG. 1 perform the following steps 206 to 208. If there is no weighting coefficient micro-grid point, the meteorological data primary estimation processing unit 124 of FIG.

  When the weighting coefficient microgrid points exist, the interactive user interface unit 13 in FIG. 1 calculates the ratio of the total number of weighting coefficient microgrid points to the total number of grid points, and the positions of the weighting coefficient microgrid points and those described later. Information such as the maximum value of each component of the weighting coefficient is displayed on the screen (step 206), and based on that, the user selects whether to automatically correct the weighting coefficient (A) or to add observation data (B) ( Step 207). When A is selected, the minute weighting factor correction processing unit 123 in FIG. 1 corrects the weighting factor for the weighting factor minute grid point according to a predetermined rule, and the weighting factor of one or more observation points is the minuteness factor. It is set so as to exceed the threshold value (ε) (step 208). This minute weighting coefficient correction processing is a portion that is the greatest feature of the embodiment, and a detailed example of the processing method will be described later. On the other hand, when B is selected, the observation data addition input unit 133 in FIG. 1 estimates the observation data of the weather observation point newly added by the user or the virtual weather observation data by some appropriate method. Then, it is read as additional input data (step 211), and the processing returns to step 201.

  Using the weighting coefficient of each observation point calculated or corrected as described above, the primary estimation value for each grid point of the weather element data is calculated by the weather data primary estimation processing unit 124 of FIG. Is calculated (step 209).

Where q (i, j, k) is the estimated value of the meteorological elements (wind velocity vector components, temperature, rainfall, etc.) at grid point (i, j, k), and q _n is the meteorological element at observation point n. The observation value, W _n (i, j, k) indicates the weighting coefficient of the observation point n with respect to the lattice point (i, j, k). When q is a component of a two-dimensional wind speed vector, as shown in Non-Patent Document 1, q _n is corrected by the difference in height between observation point n and lattice point (i, j, k). Do.

  Finally, meteorological element data at each grid point is calculated by objective analysis in which the primary estimation result obtained above is corrected and calculated based on the physical laws followed by the meteorological element data (step 210). This objective analysis is a conventional technique, and includes the mascon model widely applied as an objective analysis method for a three-dimensional wind velocity field, the technique disclosed in Patent Document 1, and the like.

  In the following, the processing content of the minute weight coefficient correction processing unit 123 of FIG.

  FIG. 3 shows an example of a detailed flow of step 208 of FIG. 2 performed by the minute weight coefficient correction processing unit 123 of FIG. First, it is determined whether or not the maximum value Wmax (i, j, k) of the weighting coefficient of each observation point with respect to the lattice point (i, j, k) is less than or equal to the minute threshold value ε (step 301). If (i, j, k) ≦ ε, the weighting factor of all observation points is ε or less, and the lattice point (i, j, k) is a weighting factor micro lattice point. At ˜307, the minute weight coefficient correction process is performed. When Wmax (i, j, k)> ε, the weighting coefficient is not corrected.

  Here, as shown in the equation (Equation 2), the weight coefficient Wn (i, j, k) of the observation point n with respect to the lattice point (i, j, k) is generally M (M ≧ 1) components W. It is expressed as a product of [m] n (i, j, k) (m = 1 to M).

  In addition, the maximum value W [m] max (i, j, k) of the m-th component of the weighting coefficient for all observation points is represented. An example of a weighting factor correction method described below is as follows. When the weighting factor of all observation points is very small, that is, equal to or less than ε, the weighting factor of all observation points is set to 1.0 as a constant, that is, weighted. It follows the rule that there is no.

  First, when the maximum value W [m] max (i, j, k) of the m-th component of the weight coefficient of the observation point n is small for all components, that is, equal to or less than ε (step 302), all the observation points The weight coefficient is set to 1.0 (step 307). Next, even when the maximum value W [m] max (i, j, k) of the m-th component of the weighting coefficient of the observation point n is not all small, that is, larger than ε (step 303), the weighting coefficient of all the observation points Is set to 1.0 (step 307). The reason for setting in this way is that although the maximum value of each component W [m] n (i, j, k) of the weighting factor is larger than ε for all the components, the total weighting factor Wn is the product of these. This is because when (i, j, k) is equal to or less than ε, the weight coefficient component cannot be selected as shown in step 304 below.

  In other cases, that is, when W [m] max (i, j, k) for each component is very small and non-small, W [m] max (i, j, k) Only for the component m in which is small (ε or less), the weighting coefficient of all observation points is set to 1.0, that is, W [m] n (i, j, k) = 1.0 (n = 1 to N) (step 304). ) This selects only the components having a value that is not very small (greater than ε) for one or more observation points from among the weighting factor components at each observation point, and the product of them is the total weighting factor Wn (i, j, k). Next, the total weighting factor Wn (i, j, k) at all observation points is again calculated as the product of each component of the weighting factor (Step 305), and the maximum value of the weighting factor is calculated. It is determined whether or not Wmax (i, j, k) is equal to or less than ε (step 306). If Wmax (i, j, k) ≦ ε, the reason is the same as the processing by the determination in step 303. The weighting factor of all observation points is set to 1.0 (step 307). When Wmax (i, j, k)> ε, Wn (i, j, k) at this time is a correction value of the total weight coefficient.

  By repeating the above processing for all grid points above the ground surface (step 308), the weight coefficient micro-grid points do not exist, and all the grid points are subjected to the meteorological element based on the weight coefficient of each weather observation point. A primary estimate can be calculated.

  (Table 1) shows the case where there are three weighting coefficient components in the minute weighting coefficient correction method described above. This is a method shown in Non-Patent Document 1, in which a primary estimation value of a wind field on a three-dimensional lattice point is obtained, a wind direction / wind speed data of a ground observation point existing on complex terrain is represented by a horizontal distance, a vertical distance, and In this method, weighted averaging is performed using a product of three kinds of weight coefficient components as a function of the terrain barrier. Here, as an empirical function of the horizontal distance r between each grid point and each observation point, the vertical distance h, and the height hb of the topographical barrier (the highest elevation point between each grid point and each observation point). Define horizontal distance weighting factor component W (r), vertical distance weighting factor component W (h) and topographic barrier weighting factor component W (hb), respectively, and the wind direction and wind speed of each observation point with respect to each grid point as their product Calculate the weighting coefficient of observation data.

  In the case of Case 1, the maximum value [W (r) ・ W (h) ・ W (hb)] max of the above three products is larger than ε and the observation point The weighting factor setting values of W (r), W (h), and W (hb) are assumed. In case 2, the maximum values of the three weighting factor components are all larger than ε, but the product is equal to or smaller than ε, so 1.0 is assigned as a constant that does not weight each weather observation point. For case 3, since all three weighting factor components are equal to or less than ε, 1.0 is used as the weighting factor as in case 2.

  In case 4, W (r) is adopted. The maximum value of W (r) exceeds the minute threshold value. Therefore, in this case, W (r) is used as the value of the weight coefficient.

  Case 5 is similar to Case 4 but uses W (h). Case 6 is similar to Case 4 and Case 5, but uses W (hb). In case 7, use W (r) and W (h). In case 8, since [W (r) · W (h)] max is smaller than the minute threshold value, a constant of 1.0 is used. In case 9, since [W (r) · W (hb)] max is larger than the minute threshold value, W (r) · W (hb) is used. In case 10, since [W (r) · W (hb)] max is smaller than the minute threshold value, 1.0 is used. In case 11, since [W (h) · W (hb)] max is larger than the minute threshold value, W (h) · W (hb) is used. In case 12, since [W (h) · W (hb)] max is smaller than the minute threshold value, 1.0 is used.

  As an example of the calculation result by the meteorological data primary estimation processing unit 124 of the meteorological data estimation apparatus of FIG. 1 described above, when the wind direction / velocity observation data of AMeDAS and the meteorological observatory is used, the three-dimensional lattice points on the complex terrain are used. FIG. 4 shows a vector display of the wind direction and the wind speed on the lattice point closest to the ground surface among the results of the primary estimation of the wind direction and the wind speed. The primary estimation method of the wind velocity field employs the method shown in Non-Patent Document 1. In FIG. 4, the calculation area is 40 km square × height 4000 m, the number of calculation grids is 20 × 20 × 50, the arrow indicates the wind speed vector, the length is proportional to the wind speed, and the direction is the leeward direction. Indicates. Wind direction and wind speed observation points are indicated by circles, and wind direction and wind speed observation data are indicated by thick arrows. In addition, the contours shown by dotted lines are the contour lines, and all the lattice points inside the area surrounded by the broken lines are the weight coefficient micro lattice points. From FIG. 4, it can be confirmed that the primary estimated value of the wind speed vector is obtained even with the weight coefficient minute grid points by the minute weight coefficient correction method shown in FIG. 3 and Table 1.

  As described above, a meteorological data estimation device that estimates meteorological element data of each grid point divided into a target space area and a grid from the meteorological element observation data of a plurality of meteorological observation points included in the target space and time domain Has the following configuration.

  The weighting factor of meteorological element data at each point corresponding to each grid point is set to each grid for at least the horizontal distance, the vertical height and the height of the topographical barrier between each grid point and each meteorological observation point, or a plurality of these. An observation data weighting coefficient calculation means and an observation data weighting coefficient calculation program for assigning weighted coefficients closer to a point and calculating their product.

  Meteorological data primary estimation processing means and weather data primary that are calculated by averaging the estimated values of meteorological element data at each grid point by multiplying the meteorological element observation data at each meteorological observation point by the observed data weighting factor calculated above. Estimation processing program.

  A minute weight coefficient judging means and a minute weight coefficient judging program for judging whether or not the maximum value of the aforementioned observation data weight coefficient at each meteorological observation point is not more than a preset minute threshold value for each grid point.

  When it is determined that the calculated maximum value of the observation data weighting factor is less than or equal to the minute threshold value, do not weight one or more of the observation data weighting factor components whose maximum value is less than or equal to the minute threshold value. A minute weight coefficient correction processing means and a minute weight coefficient correction processing program for correcting and setting a product of each component so as to exceed a minute threshold value by giving a constant.

  A modified weather data estimation processing means and a modified weather data estimation processing program for correcting the weather data estimated by the weather data primary estimation processing using the corrected minute weight coefficient.

  Next, as an example of the calculation result by the meteorological data objective analysis processing unit 125 of the meteorological data estimation apparatus of FIG. 1, based on the primary estimation result of the wind speed vector shown in FIG. FIG. 5 shows a vector display of the wind direction and wind speed on the lattice point closest to the ground surface among the objective analysis results corrected as follows. From FIG. 5, it can be confirmed that the estimated wind velocity field in the weight coefficient micro lattice point region is improved from the viewpoint of continuity with the surrounding wind velocity field as compared with FIG. 4.

  In addition, an Example is not limited to the said embodiment, In the implementation stage, in the range which does not deviate from the characteristic of an Example, a structure means, a structure step, or a structure procedure can be changed and embodied. For example, another method may be employed as a minute weight coefficient correction rule, or the weather data estimation apparatus may be configured by deleting the interactive user interface unit and the weather data objective analysis processing unit.

The figure which shows one Example of a structure of a weather data estimation apparatus. The figure which shows one Example of the processing flow in a weather data estimation apparatus. The figure which shows one Example of a minute weighting coefficient correction process flow. The figure which shows one Example of the weather data primary estimation process by a weather data estimation apparatus. The figure which shows one Example of the weather data objective analysis process by a weather data estimation apparatus.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Input device, 2 ... Output device, 3 ... Memory | storage device, 10 ... Arithmetic processing device, 11 ... Data setting part, 12 ... Weather data estimation calculation part, 13 ... Interactive user interface part, 14 ... Weather data estimation result display 122, a minute weighting coefficient determination unit, 123, a minute weighting coefficient correction processing unit.

Claims (3)

In the meteorological data estimation device for estimating the meteorological element data of each grid point obtained by dividing the target space area into a grid from the meteorological element observation data of a plurality of meteorological observation points included in the target space and time domain,
The weighting factor of meteorological element observation data at each point corresponding to each grid point is set to at least the horizontal distance, vertical height and height of topographic barrier between each grid point and each meteorological observation point, or a plurality of these An observation data weighting factor calculation means that assigns a weighted coefficient (weighting factor component) closer to each grid point and calculates the product of those components, and an estimate of meteorological element data at each grid point Meteorological data primary estimation processing means that is calculated by multiplying the meteorological element observation data of the observation point by the above-mentioned calculated observation data weighting factor, and the maximum of the aforementioned observation data weighting factor of each meteorological observation point for each grid point A minute weight coefficient judging means for judging whether the value is less than or equal to a preset minute threshold value, and that the calculated maximum value of the observed data weight coefficient is less than or equal to the minute threshold value. The weight of each weather station by using the product of only the observation data weight coefficient components whose maximum value exceeds the minute threshold, or by assigning a constant without weighting all the stations. A minute weight coefficient correction processing means for correcting and setting so that the maximum value of the coefficient exceeds the minute threshold value, and using the corrected weight coefficient, the weather data estimated by the weather data primary estimation process is corrected. A meteorological data estimation processing means, and a weather data estimation device. Meteorological data used in a meteorological data estimation device that estimates meteorological element data at each grid point by dividing the target space area into a grid from meteorological element observation data at multiple meteorological observation points included in the target space and time domain In the estimation program,
The weighting factor of meteorological element observation data at each point corresponding to each grid point is set to at least the horizontal distance, vertical height and height of topographic barrier between each grid point and each meteorological observation point, or a plurality of these An observation data weighting coefficient calculation program that assigns weighted coefficients (weighting coefficient components) closer to each grid point and calculates the product of those components, and estimates of meteorological element data at each grid point The meteorological data primary estimation processing program that is calculated by multiplying the meteorological element observation data of the observation point by the above-mentioned calculated observation data weighting factor, and the maximum of the aforementioned observation data weighting factor of each meteorological observation point for each grid point A small weighting coefficient determination program for determining whether the value is less than or equal to a predetermined small threshold, and the maximum value of the calculated observation data weighting coefficient is a small threshold When it is determined that the value is less than or equal to the value, use the product of only the observation data weight coefficient components whose maximum value exceeds the minute threshold, or assign a constant without weighting all observation points, A fine weighting factor correction processing program that corrects and sets a maximum value of the weighting factor of a weather observation point so as to exceed a fine threshold value, and is estimated by the weather data primary estimation processing using the corrected weighting factor. A weather data estimation program for use in a weather data estimation apparatus, comprising: a modified weather data estimation processing program for correcting weather data. A meteorological data estimation method using a meteorological data estimation device for estimating meteorological element data of each grid point obtained by dividing the target space area into a grid from meteorological element observation data of a plurality of meteorological observation points included in the target space and time domain In
Using the observation data weighting factor calculation means, the weighting factor of the meteorological element observation data at each point corresponding to each grid point is set as any of the horizontal distance, vertical height, and barrier height between each grid point and each meteorological observation point. Or a coefficient (weighting coefficient component) weighted as it is closer to each grid point for a plurality of these, and calculated as the product of those components, and meteorological elements at each grid point by the weather data primary estimation processing means The estimated value of the data is calculated by multiplying the meteorological element observation data of each meteorological observation point by the observation data weighting coefficient calculated above, and averaging the result. It is determined whether the maximum value of the observation data weighting factor described above is less than or equal to a preset minute threshold value, and the calculated observation data is processed by the minute weighting factor correction processing means. Use the product of only the observation data weighting coefficient components whose maximum value exceeds the minute threshold, or weight all observation points. Without changing the maximum value of the weight coefficient of each weather observation point so that it exceeds the minute threshold value, and using the corrected weight data estimation processing means, A weather data estimation method by a weather data estimation device, wherein the weather data estimated by the weather data primary estimation process is corrected.

Images