JP4595078B2 - Apparatus and method for estimating three-dimensional distribution of rainfall intensity and amount of rainwater - Google Patents

Description

The present invention estimates the three-dimensional distribution of the rainfall intensity and the amount of rain water based on the phase difference K _DP , the reflection factor difference Z _DR and the reflection factor Z _H obtained by the multi-parameter radar. The present invention relates to a distribution estimation method and apparatus.

Knowing the three-dimensional distribution of rainfall intensity and rainfall, and the two-dimensional distribution of vertical accumulated rainwater is useful for investigating the structure of the precipitation system and the mechanism of its occurrence and development, and for short-term prediction of heavy rain that may cause disasters. is important. Weather radar is an effective observation device for such purposes. The Ministry of Land, Infrastructure, Transport and Tourism has installed about 40 conventional radars, called conventional radars, throughout the country to monitor and forecast rainfall. In this radar system, by emitting one type of radio wave from the antenna to the atmosphere, the emitted radio wave hits the rain and returns to the radar antenna. Therefore, the intensity of the radio wave is measured and the position of the rain is determined from that time. The rainfall intensity R and the amount of rainwater M are estimated from information called the reflection factor Z. The three-dimensional distribution of the rainfall intensity and the amount of rainwater can be obtained by observing while changing the elevation angle of the antenna. In that case, an empirical formula called a ZR relational expression or a ZM relational expression is used. It is well known that this relational equation is difficult to make accurate measurements due to various error factors such as radar hardware calibration, rain type dependent, and rain attenuation effects. Yes. A method for estimating the rainfall intensity and the amount of rainwater using a polarization radar has been proposed in contrast to the method for estimating the rainfall intensity and the amount of rainwater using a conventional radar (see, for example, Patent Documents 1 and 2). In these methods, by using the reflection factor obtained from the polarization radar Z _H and differential reflectivity factor Z _DR, it is intended to accurately estimate the rainfall intensity and rainwater volume. The estimation method of the rainfall intensity and the amount of rainwater using the reflection factor Z _H and the reflection factor difference Z _DR obtained from the polarization radar is theoretically a highly accurate estimation method. Since the factor Z _H and the reflection factor difference Z _DR are affected by attenuation due to rain, measurement in heavy rain may be accompanied by a large error as in the case of the conventional radar.
Japanese Patent No. 2660942 JP 2003-344556 A

The estimation method of the rainfall intensity and the amount of rainwater using the conventional radar is not good in estimating the rain intensity due to various error factors such as radar calibration error, rain attenuation, and fluctuation of raindrop particle size distribution. Therefore, the Japan Meteorological Agency and others are trying to improve the accuracy by correcting the estimated value with the data of the rain gauge installed on the ground, but it is accurate in mountain areas where there are not enough rain gauges available or at sea where there are no rain gauges available. There is a disadvantage that improvement cannot be expected. On the other hand, the estimation method of the rainfall intensity and the amount of rain water using the reflection factor Z _H and the reflection factor difference Z _DR obtained from the polarization radar is theoretically a highly accurate estimation method. Since the reflection factor Z _H and the reflection factor difference Z _DR are affected by attenuation due to rain, there is a large error in estimating the rainfall intensity and the amount of rain water during heavy rain. In addition, since the conventional estimation formula does not consider the influence of temperature and observation elevation angle, an error occurs when obtaining the three-dimensional distribution of rainfall intensity and rainwater volume.

  The present invention solves the above-described problems, and makes it possible to accurately estimate the three-dimensional distribution of rainfall intensity and amount of rainwater.

For this purpose, the present invention estimates the three-dimensional distribution of rainfall intensity and rainfall based on the phase difference K _DP , the reflection factor difference Z _DR , and the reflection factor Z _H obtained by the multi-parameter radar. As a 3D distribution estimation method,

Either of the above is used as an estimation formula for rainfall intensity,

Is one of the formulas for estimating the amount of rainwater,
Temperature profile t (r, θ) in the range direction from the ground temperature t ₀ , radar range r, observation elevation angle θ, and standard atmospheric temperature decrease rate Γ (= 0.065 ° C./m)

, And the coefficients and power exponents a _i (i = 1, 2), a _{i ′} (i = 1, 2), b _i (i = 1) of the above estimation equations in consideration of temperature dependency and elevation angle dependency 2), b _{i ′} (i = 1, 2), c _i (i = 1, 2, 3), c _{i ′} (i = 1, 2, 3), d _i (i = 1, 2, 3) ), D _{i ′} (i = 1, 2, 3) ,
Based on the selection of the calculation mode, the presence or absence of attenuation correction is determined to correct the rain attenuation. Based on the selection of the calculation method, the estimation formula of [Formula 1] is determined to calculate the rainfall intensity. It is characterized in that the estimation formula of [Equation 2] is determined and the amount of rainwater is calculated to estimate the three-dimensional distribution of the rainfall intensity and the amount of rainwater.

As a three-dimensional distribution estimator for rainfall intensity and rainfall, which estimates the three-dimensional distribution of rainfall intensity and rainfall based on phase difference K _DP , reflection factor difference Z _DR and reflection factor Z _H obtained by multi-parameter radar ,
A selection means for selecting a calculation mode and a calculation method;
Correction means for correcting the rain attenuation by the self-consistent method using the information on the polarization phase difference φ _{DP that} is not affected by the rain attenuation,
Temperature profile t (r, θ) in the range direction from the ground temperature t ₀ , radar range r, observation elevation angle θ, and standard atmospheric temperature decrease rate Γ (= 0.065 ° C./m)

A temperature range profile calculation means for calculating
The specific polarization phase difference K _DP , the reflection factor difference Z _DR , the reflection factor Z _H , the elevation angle θ, and the temperature profile t (r, θ) in the range direction are input and based on the calculation mode selected by the selection means. The rain attenuation is corrected by the correction means and based on the selected calculation method.

Either of the above is used as an estimation formula for rainfall intensity,

Is one of the formulas for estimating the amount of rainwater,
Estimation means for estimating a three-dimensional distribution of the rainfall intensity and the amount of rainwater, and the estimation means includes a coefficient and a power index a _i (i = 1, 1) of each estimation formula taking temperature dependency and elevation angle dependency into consideration. 2), a _{i ′} (i = 1, 2), b _i (i = 1, 2), b _{i ′} (i = 1, 2), c _i (i = 1, 2, 3), c _{i ′} (I = 1, 2, 3), d _i (i = 1, 2, 3), d _{i ′} (i = 1, 2, 3) are used.

In addition, the estimation means may use the K _DP method for calculating the R (K _DP ) and M (K _DP ) on the condition that the calculation mode is no attenuation correction, R (Z _H ) or R (K _DP ), M (Z _H ) or M (K _DP ) is selected from the composite method A, and R (Z _H ) or R ( _{K DP), M (Z H} ) or M (K _DP) composite method B perform _{computations, R (K DP, Z DR} ), M (K DP, the calculation of _{Z DR) (K DP, Z} DR ) _{method, R (Z H, Z DR} ), M (Z H, calculates the Z _DR) (Z _H, and selects one of Z _DR) method.

  According to the present invention, a 3 cm wavelength multi-parameter radar is used to estimate the three-dimensional distribution of the rainfall intensity and the amount of rainwater based on the rainfall intensity estimation formula taking into account the observation elevation angle and temperature change, and the rainwater quantity estimation formula. It can be greatly improved.

  Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram for explaining an embodiment of a three-dimensional distribution estimation apparatus for rainfall intensity and rainwater according to the present invention, and FIG. 2 is a flowchart for explaining a method for estimating rainfall intensity and rainwater.

In FIG. 1, 1 is a multi-parameter radar system, 2 is measurement data, 3 is a three-dimensional distribution measurement system for rainfall intensity and rainfall, 4 is an estimation unit, 5 is an attenuation correction unit, and 6 is 3 of rainfall intensity and rainfall. A dimension distribution data storage unit, 7 is an input parameter processing unit, 8 is a profile calculation unit in the temperature range direction, and 9 is an estimation calculation unit. The multi-parameter radar system 1 uses a multi-parameter radar system with a wavelength of 3 cm and uses two types of radio waves, horizontal polarization and vertical polarization. When radio waves are transmitted in the rain, the phase velocity of horizontal polarization is slightly slower than that of vertical polarization. This is called an inter-polarization phase difference, and this embodiment uses this inter-polarization phase difference information. The method using the phase difference information between polarizations is highly accurate because it is not easily affected by various error factors such as the disadvantages of conventional radar, that is, radar calibration error, rain attenuation, and raindrop size distribution fluctuation. Rain information can be estimated. The measurement data 2 includes parameters such as the specific polarization phase difference K _DP , the reflection factor difference Z _DR , the reflection factor Z _H , the observation elevation angle (antenna altitude angle) θ, and the like, which are output from the multi-parameter radar system 1. It estimates the three-dimensional distribution of rainwater. The rainfall intensity and rainwater volume three-dimensional distribution estimation system 3 includes an estimation unit 4, an attenuation correction unit 5, and a rainfall intensity and rainwater volume three-dimensional distribution data storage unit 6. The estimation unit 4 includes an input parameter processing unit 7, It consists of a profile calculation unit 8 and an estimation calculation unit 9 in the temperature range direction.

Next, a method for estimating the three-dimensional distribution of the high intensity and the rainwater amount by the three-dimensional distribution estimation system 3 for the rainfall intensity and the rainwater amount will be described. As shown in FIG. 2, first, the calculation mode (with or without attenuation correction) and calculation method (K _DP method, composite method A, composite B, (K _DP , Z _DR ) method, (Z _DR , Z _H ) method) Selective input is performed (step S1), and specific polarization phase difference K _DP , reflection factor difference Z _DR , reflection factor Z _H , and elevation angle θ are input as observation data from the multi-parameter radar system (step S2). The temperature t ₀ is input (step 3), the temperature profile in the range direction is calculated (step 4), and the temperature-dependent temperature at each point is obtained. Next, it is determined whether the calculation mode has no attenuation correction or has attenuation correction (step 5). When the decision in step 5 is no attenuation correction, the calculation method or K _DP method performs Kano determination Composite Method A (step S6), and if K _DP method based on the determination, and rain by the composite method A Calculate the three-dimensional distribution of rainwater. The _KDP method or the composite method A is selected when real-time performance is required and faster calculation is required. If the determination in step S5 is attenuation correction, after performing rain attenuation correction processing (step 7), whether the calculation method is the composite method B, the (K _DP , Z _DR ) method, or (Z _DR , Z _H ) method is determined (step S8), and based on the determination, the rainfall intensity and the amount of rainwater by composite method B, (K _DP , Z _DR ) method or (Z _DR , Z _H ) method are determined. Calculate the three-dimensional distribution.

  In this embodiment, as an estimation formula of rainfall intensity,

To calculate the rainwater amount,

Do one of the calculations. The coefficients and power exponents a _i (i = 1, 2), a _{i ′} (i = 1, 2), b _i (i = 1, 2), b _{i ′} (i = 1, 2), c _i (i = 1, 2, 3), c _{i ′} (i = 1, 2, 3), d _i (i = 1, 2, 3), d _{i ′} (i = 1, 2, 3) are Considering temperature dependency and elevation angle dependency.

Coefficients and exponents a _i (i = 1, 2), a _{i ′} (i = 1, 2), b _i (i = 1, 2), b _{i ′} (i = 1, 2), c _i (i = 1, 2, 3), c _{i ′} (i = 1, 2, 3), d _i (i = 1, 2, 3), d _{i ′} (i = 1, 2, 3) are raindrop particle size distributions. It can be obtained by scattering calculation based on raindrop particle size distribution data actually measured by a measuring device. For the calculation of scattering, the T-matrix method is used. As the calculation conditions of scattering, for example, there are three types of temperature (0, 15, 30 ° C.) and five types of antenna altitude angles (0, 10, 20, 30, 40). °) is used. By the way, in the estimation formula of the rainfall intensity and the amount of rain water by the conventional polarization radar, the elevation angle of the coefficient and the change of the temperature are not taken into consideration. As in this embodiment, the three-dimensional distribution of precipitation intensity and amount of rainwater can be obtained with high accuracy by taking temperature dependency and elevation angle dependency into consideration. Furthermore, each calculation content is explained in full detail.

  FIG. 3 is a diagram for explaining the calculation of the temperature profile in the range direction. Assuming standard atmosphere, the vertical profile of temperature is

It can be obtained by calculating Here, Γ is the temperature reduction rate of the standard atmosphere (0.065 ° C./m), and z is the height from the ground.

On the other hand, since z = r sin θ, the temperature profile in the range direction is calculated by inputting the radar range r, the observation elevation angle θ, and the temperature t ₀ near the ground.

It can be obtained by calculating

FIG. 4 is a flowchart for explaining the _KDP method. As shown in FIG. 4, the calculation of the three-dimensional distribution of rainfall intensity and amount of rainwater by the K _DP method starts with parameters α _i (i = 0, 1, 2, 3), β _i (i = 0, 1) After calculating α _i ′ (i = 0, 1, 2, 3) and β _i ′ (i = 0, 1) (step S11), the coefficient should be determined from these parameters and the antenna elevation angle θ. index _{b i (i = 1,2),} b i ' perform calculations of (i = 1, 2) (step 12), the parameters and the temperature t, rainfall intensity based on the elevation angle θ R (K _DP) and rainwater amount M (K _DP ) is calculated (step S13). The K _DP method, since using the K _DP method is measured from the radar, it is possible to perform the fastest calculation, is an effective means in the case which requires real time. However, since the estimation accuracy in the case of light rain deteriorates, caution is required when the rainfall intensity is 8 mm / h or less as a guide.

In the K _DP method for performing calculations, coefficients and exponents _{b i (i = 1,2),} b i '(i = 1,2) is dependent on the observation antenna elevation θ and the temperature t, b _i ( i = 1, 2), b _{i ′} (i = 1, 2),

Is assumed to be a polynomial. Here, it is assumed that the parameters α _i , α _i ′ (i = 0, 1, 2, 3), β _i , β _i ′ (i = 0, 1) are expressed by a quadratic function of the temperature t. The quasi-Newton method (sequential quadratic programming method) is applied to the scattering simulation result based on the raindrop size distribution data.

  From the above formulas [Equation 5] and [Equation 6], the rainfall intensity and the amount of rainwater at an arbitrary elevation angle θ and temperature t are calculated.

FIG. 5 is a flowchart for explaining the composite method A. As shown in FIG. 5, the calculation of the three-dimensional distribution of the rainfall intensity and the amount of rainwater by the composite method A starts with R (Z _H ) and R depending on whether the observed K _DP value is larger than K _DP _thr. Which estimation formula (K _DP ) is used is selected (step S21). When the value of K _DP is smaller than K _DP _thr the parameters for estimation formula R (Z _H) from the temperature _{t α i (i = 0,1,2,3)} , β i (i = 0,1), α _i ′ (i = 0, 1, 2, 3) and β _i ′ (i = 0, 1) are obtained (step S22), and a coefficient and an exponent a _i (i = 1) are calculated from these parameters and the antenna elevation angle θ. , 2), a _{i ′} (i = 1, 2) is obtained (step S23), and the rainfall intensity is calculated using the estimation formula R (Z _H ) in [Equation 1]. Further, the amount of rainwater is calculated using the estimation formula M (Z _H ) in [Equation 2] (step S24). When the value of K _DP is greater than K _DP _thr the parameters for estimation formula R (K _DP) from the temperature _{t α i (i = 0,1,2,3)} , β i (i = 0,1), α _i ′ (i = 0, 1, 2, 3), β _i ′ (i = 0, 1) are obtained (step S25), and the coefficient and power exponent a _i (i = 1) are determined from these parameters and the antenna elevation angle θ. , 2), a _{i ′} (i = 1, 2) are obtained (step S26), and the rainfall intensity is estimated using the estimation formula R (K _DP ) in [Equation 1] in [Equation 2]. The amount of rainwater is calculated using the equation M (K _DP ) (step S27). The composite method A uses the classical estimation formulas R (Z _H ) and M (Z _H ) for light rain, and the weak point of the K _DP method described above, that is, the K _DP method estimates for light rain. This is to improve the point that accuracy decreases. However, the reflection factor Z _H used here is not corrected for rain attenuation. This is because processing is performed in real time.

If the rain attenuation is corrected, it is possible to estimate the rain intensity and the amount of rain water with higher accuracy. Polarization radar can correct the rain attenuation of Z _HH and Z _DR by using the information of the phase difference φ _DP between the polarization waves that is not affected by the rain attenuation. In the flowchart of FIG. 2, the rain attenuation correction method of the polarization parameter used in the rain attenuation correction (step S7) is a self-consistent method. Bringi et al. (2001) The one used for the radar is improved for the X band. In the self-consistent method, “the integrated value of the attenuation coefficient A _HH from the range r ₁ to r ₀ agrees with the polarization phase difference Δφ _DP = φ _DP (r ₀ ) −φ _DP (r ₁ ) in the same section”. The optimum value is obtained using the binding condition. Furthermore, an optimum value is obtained using the constraint that “Z _DR at the far end (r ₀ ) of the rain echo is expressed by a linear function of Z _H ” (Reference 1: Bringi, VN, TD Keenan, and V. Chandrasekar, 2001: Correcting C-band radar reflectivity and differential reflectivity data for rain attenuation: A self-consistent method with constraints. IEEE Trans. Geosci. Remote Sens., 39, 1906-1915.).

FIG. 6 is a flowchart for explaining the composite method B. Calculation of the three-dimensional distribution of the rainfall intensity and the amount of rainwater by the composite method B is similar to the composite method A, but the selection conditions for the estimation formula are different. Further, the composite method A uses a reflection factor Z _H without rain attenuation correction, whereas the composite method B uses Z _{H —} cor corrected for rain attenuation. As shown in FIG. 6, first, R (Z _H depends on whether the observed K _DP value is larger than K _{DP —} thr and whether the attenuation corrected Z _{H —} cor value is larger than Z _{H —} thr. ) Or R (K _DP ) to be used is selected (step S31). K _DP of when the value is the value of time or Z _H _cor follows K _DP _thr is below Z _H _thr the parameters for estimation formula R (Z) from the temperature t α _{i (i} = 0,1,2,3) , Β _i (i = 0, 1), α _i ′ (i = 0, 1, 2, 3), β _i ′ (i = 0, 1) are obtained (step S32), and these parameters and the antenna elevation angle θ From the coefficients and power exponents a _i (i = 1, 2) and a _{i ′} (i = 1, 2) (step S33), and using the estimation formula R (Z _H ) of [Equation 1] The calculation is performed, and the amount of rainwater is calculated using the estimation formula M (Z _H ) of [Equation 2] (step S34). K When greater than when a and Z _H value of _cor is Z _H _ _thr value is greater than K _DP _ _thr the _DP, the parameter α _{i (i} = 0 for estimation formula R (K _DP) from the temperature t, 1, 2, 3), β _i (i = 0, 1), α _i ′ (i = 0, 1, 2, 3), β _i ′ (i = 0, 1) are obtained (step S35). And the exponents a _i (i = 1, 2) and a _{i ′} (i = 1, 2) are obtained from the parameters of the antenna and the antenna elevation angle θ (step S36), and the estimation formula R (K _DP ) of [Equation 1] Is used to calculate the rainfall intensity, and the rainwater quantity is calculated using the estimation formula M (K _DP ) of [Equation 2].

FIG. 7 is a flowchart for explaining the (K _DP , Z _DR ) method. As shown in FIG. 5, the calculation of the three-dimensional distribution of the rainfall intensity and the amount of rainwater by the (K _DP , Z _DR ) method starts with the parameters α _i (i = 0, 1, 2, 3), β _i from the temperature t. (I = 0, 1), α _i ′ (i = 0, 1, 2, 3), β _i ′ (i = 0, 1) are obtained (step S41), and a coefficient is calculated from these parameters and the antenna elevation angle θ. The power exponents c _i (i = 1, 2) and c _{i ′} (i = 1, 2) are obtained (step S42), and the rainfall intensity is calculated using the estimation formula R (K _DP , Z _DR ) of [Equation 1]. Calculation is performed, and the amount of rainwater is calculated using the estimation formula M (K _DP , Z _DR ) of [Equation 2] (step S43). The Z _DR used here is after rain attenuation correction. Although K _{DP — thr} = 0.3 ° / km is assumed, a slightly larger value may be used depending on the situation. K _{DP — thr} = 0.3 ° / km is 7.3 mm / h in terms of rainfall intensity when the elevation angle is 0 ° and the temperature is 15 ° C.

FIG. 8 is a flowchart for explaining the (Z _H , Z _DR ) method. As shown in FIG. 8, the calculation of the three-dimensional distribution of the rainfall intensity and the amount of rainwater by the (Z _H , Z _DR ) method starts with parameters α _i (i = 0, 1, 2, 3), β _i from the temperature t. (I = 0, 1), α _i ′ (i = 0, 1, 2, 3), β _i ′ (i = 0, 1) are obtained (step S41), and a coefficient is calculated from these parameters and the antenna elevation angle θ. The power indices c _i (i = 1, 2) and c _{i ′} (i = 1, 2) are obtained (step S42), and the rainfall intensity is calculated using the estimation formula R (Z _H , Z _DR ) of [Equation 1]. Calculation is performed, and the amount of rainwater is calculated using the estimation formula M (M _{H —} cor, Z _{DR —} cor) of [Equation 2] (step S43). Note that Z _H and Z _DR used here are after rain attenuation correction.

[Expression 1] Rain intensity estimation formula, coefficient and exponents a _i (i = 1, 2), b _i (i = 1, 2), c _i (i = 1, 2, 3), d _i (i = 1, 2, 3) and a rain water quantity estimation formula of [Equation 2] a _{i ′} (i = 1, 2), b _{i ′} (i = 1, 2), c _{i ′} (i = 1, 2, 3) and d _{i ′} (i = 1, 2, 3) have already been proposed in the literature, but their values do not take into account changes in elevation angle or temperature. Therefore, an error may occur when the three-dimensional distribution of the rainfall intensity and the amount of rainwater is obtained using the value.

  The method used in this case takes into account the temperature dependence and elevation angle dependence of these coefficients and power exponents, and is inventive in this respect. That is, in order to accurately measure the three-dimensional distribution of the rainfall intensity and the amount of rainwater in the present invention, the coefficient and index of the estimation formula of the rain intensity and the rainwater amount of [Equation 1] and [Equation 2] are measured with the temperature t. One feature is that it is a function of the elevation angle θ. Hereinafter, it will be described in detail how much error occurs when the influence of the temperature t and the observation elevation angle θ is ignored.

First, the results on precipitation intensity will be described. A relative error that occurs when a coefficient of θ = 0 ° and a power index are used for other θ is ΔRθ / R ₀ . Further, a relative error that occurs when a coefficient of t ₀ = 20 ° C. and a power index are used for another temperature t is represented by ΔR _t / R ₀ . The results obtained for each estimation formula are shown below.

(A) The results of the elevation angle dependency and the temperature dependency in the case of the estimation formula R (Z _H ) are shown in FIG. As can be seen from FIG. 9A, the relative error that occurs when the elevation angle θ is a coefficient of 0 ° and the exponent is used for other elevation angles increases with the elevation angle, but the magnitude is 3% even when θ = 60 °. Degree. On the other hand, as can be seen from FIG. 9B, the relative miscalculation that occurs when the coefficient and power index when the temperature is 20 ° C. is used for other temperatures increases as the temperature decreases, and increases as the rainfall intensity increases. . For example, at t = 0 ° C., when the rainfall intensity R ₀ = 160 mm / h, the relative error is about −10%.

(B) FIG. 10 shows the results of the elevation angle dependency and the temperature dependency in the case of the estimation formula R (K _DP ). As can be seen from FIG. 10A, the absolute value of the relative error generated when the coefficient at the elevation angle of 0 ° is used for other elevation angles increases with the elevation angle, and underestimation of about 10% at θ = 20 °, θ = 60 ° results in an underestimation of about 60%. When θ = 5 ° or less, the absolute value of the relative error is about 1% or less, and the elevation angle dependency can be ignored. Such elevation angle dependency is due to the elevation angle dependency of the coefficient b ₁ . On the other hand, as can be seen from FIG. 10B, the relative miscalculation that occurs when the coefficient and power index when the temperature is 20 ° C. is used for other temperatures is about ± 2%. When R ₀ is larger than 40 mm / h, overestimation is performed, and when R ₀ is smaller than 40 mm / h, it is underestimated. ΔR _t / R ₀ is about 2% at most and practically the temperature dependence can be ignored.

(C) FIG. 11 shows the results of the elevation angle dependency and the temperature dependency in the case of the estimation formula R (K _DP , Z _DR ). As can be seen from FIG. 11A, the absolute value of ΔRθ / R ₀ increases with the elevation angle. For example, when R ₀ is 40 mm / h, about 7% underestimation occurs when θ = 20 °, and about 55% underestimation occurs when θ = 60 °. Such elevation angle dependency of the relative error is due to the elevation angle dependency of the coefficients c ₁ and c ₃ and the magnitude of Z _DR . On the other hand, as can be seen from FIG. 11B, the relative miscalculation that occurs when the coefficient and power exponent when the temperature is 20 ° C. is used for other temperatures is about −1% to + 3%. The result is very similar to R (K _DP ), overestimating when R ₀ is greater than 20 mm / h and underestimating when less than 20 mm / h.

(D) FIG. 12 shows the results of elevation angle dependency and temperature dependency in the case of the estimation formula R (Z _H , Z _DR ). As can be seen from FIG. 12A, ΔRθ / R ₀ increases with the elevation angle. For example, when R ₀ = 40 mm / h and θ = 20 °, an overestimation of about 20% is brought about, and when θ = 60 °, an overestimation of about 160% is brought about. When R ₀ = 160 mm / h, overestimation of about 30% is obtained at θ = 20 °, and overestimation of about 250% is obtained at θ = 60 °. The elevation angle dependency of ΔRθ / R _{0 is due to} the elevation angle dependency of the coefficient d ₁ and the power index d ₃ and does not depend on the magnitude of Z _H. On the other hand, as can be seen from FIG. 12B, the relative miscalculation that occurs when the coefficient when the temperature is 20 ° C. is used for other temperatures is that R ₀ = 10 mm / h to 160 mm / h until θ is 30 °. It is overestimated in the range of 9% to 3%. It becomes slightly larger at θ = 60 °, resulting in an overestimation of 10% to 6% in the range of R ₀ = 10 mm / h to 160 mm / h. The degree of overestimation decreases as the rainfall intensity R ₀ increases.

Coefficients and power indices a _i (i = 1, 2), b _i (i = 1, 2), c _i (i = 1, 2, 3), d _i ( The elevation angle dependency and temperature dependency of i = 1, 2, 3) are the coefficient of the rainwater estimation equation and the power exponents a _{i ′} (i = 1, 2), b _{i ′} (i = 1, 2), It can be shown from the result of the scattering simulation that c _{i ′} (i = 1, 2, 3) and d _{i ′} (i = 1, 2, 3) are also present (not shown).

  Therefore, in order to accurately estimate the three-dimensional distribution of rainfall intensity and rainfall, the coefficients and power exponents of the estimation equations in [Equation 1] and [Equation 2] take temperature dependence and elevation angle dependence into consideration. There must be.

  Below, the example which applied this invention about actual rainfall is shown. FIG. 13 is an external view and system configuration diagram of a 3 cm wavelength multi-parameter radar to which the present invention is applied. This radar is a multi-parameter developed and completed in 2000 by the National Research Institute for Earth Science and Disaster Prevention. The white spherical cover in the photo is a radome with a parabolic antenna inside, and various devices are stored in a container under the antenna mount. A characteristic of this radar is that it has a wavelength of 3 cm, emits two types of radio waves of horizontal polarization and vertical polarization, and measures various polarization parameters.

FIG. 14 is a diagram showing the measurement principle of a conventional radar and the measurement principle of the 3 cm wavelength multi-parameter radar system apparatus used in the present invention. FIG. 15 is a parameter measured by the 3 cm wavelength multi-parameter radar system apparatus used in the present invention. It is an example. While the conventional radar measures the reflection factor Z _H , the multi-parameter measures the reflection factor difference Z _DR and the specific polarization phase difference K _DP in addition to Z _H.

  FIG. 16 is an example of a three-dimensional distribution of rainfall intensity obtained by the present invention (Typhoon No. 10 on August 9, 2003), and FIG. 17 is an example of a three-dimensional distribution of rainwater obtained by the present invention (2003). August 9th, Typhoon No.10).

  Note that the results of the observation elevation angle dependency and temperature dependency of each estimation formula of the rainfall intensity and the amount of rain water exemplified in the present invention relate to the multi-parameter radar of 3 cm wavelength, but the multi-parameter radar of other wavelengths also applies. Similar discussions are possible.

FIG. The flowchart of the estimation method of rainfall intensity and the amount of rainwater. The flowchart of the calculation method of the temperature profile of a range direction. The flowchart of a calculation method ( _KDP method). The flowchart of the calculation method (composite method A). The flowchart of the calculation method (composite method B). The flowchart of the calculation method (( _KDP , _ZDR ) method). Flow calculation method ((Z _DR, Z _H) method). (A): R (Z _H ) relative error (t = 20 ° C.) generated when the coefficient and power exponent when the antenna elevation angle is 0 ° are used for other elevation angles. (B): R (Z _H ) relative error (θ = 5 °) generated when the coefficient and power exponent at a temperature of 20 ° C. are used for other temperatures. (A): R (K _DP ) relative error (t = 20 ° C.) generated when the coefficient and power exponent when the antenna elevation angle is 0 ° are used for other elevation angles. (B): R (K _DP ) relative error (θ = 5 °) generated when the coefficient and power exponent at a temperature of 20 ° C. are used for other temperatures. (A): R (K _DP , Z _DR ) relative error (t = 20 ° C.) that occurs when the coefficient and exponent when the antenna elevation angle is 0 ° are used for other elevation angles. (B): R (K _DP , Z _DR ) relative error (θ = 5 °) that occurs when the coefficient and power exponent at a temperature of 20 ° C. are used for other temperatures. (A): R (Z _H , Z _DR ) relative error (t = 20 ° C.) that occurs when the coefficient and exponent when the antenna elevation angle is 0 ° are used for other elevation angles. (A): R (Z _H , Z _DR ) relative error (θ = 5 °) generated when the coefficient and power exponent at a temperature of 20 ° C. are used for other temperatures. The external appearance and hardware block diagram of the 3cm wavelength multiparameter radar system apparatus used for this invention. Measurement principle of conventional radar and measurement principle of 3 cm wavelength multi-parameter radar system apparatus used in the present invention. The example of the parameter measured by the 3cm wavelength multiparameter radar system apparatus used for this invention. An example of a three-dimensional distribution of rainfall intensity obtained by the present invention (August 9, 2003, Typhoon No. 10). The example of the three-dimensional distribution of the amount of rainwater calculated | required by this invention (August 9, 2003, typhoon No. 10).

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Multi-parameter radar system, 2 ... Measurement data, 3 ... Three-dimensional distribution measurement system of rainfall intensity and rain water quantity, 4 ... Estimation part, 5 ... Attenuation correction part, 6 ... Three-dimensional distribution data storage of rainfall intensity and rain water quantity Unit 7 input parameter processing unit 8 temperature profile calculation unit in temperature range direction 9 estimation calculation unit

Claims (4)

A method for estimating the three-dimensional distribution of rainfall intensity and rainfall based on the phase difference K _DP , the reflection factor difference Z _DR and the reflection factor Z _H obtained by multi-parameter radar. There,
Either of the above is used as an estimation formula for rainfall intensity,
Is one of the formulas for estimating the amount of rainwater,
Temperature profile t (r, θ) in the range direction from the ground temperature t ₀ , radar range r, observation elevation angle θ, and standard atmospheric temperature decrease rate Γ (= 0.065 ° C./m)
, And the coefficients and power exponents a _i (i = 1, 2), a _{i ′} (i = 1, 2), b _i (i = 1) of the above estimation equations in consideration of temperature dependency and elevation angle dependency 2), b _{i ′} (i = 1, 2), c _i (i = 1, 2, 3), c _{i ′} (i = 1, 2, 3), d _i (i = 1, 2, 3) ), D _{i ′} (i = 1, 2, 3) ,
Based on the selection of the calculation mode, the presence or absence of attenuation correction is determined to correct the rain attenuation. Based on the selection of the calculation method, the estimation formula of [Formula 1] is determined to calculate the rainfall intensity. A method for estimating a three-dimensional distribution of rainfall intensity and rainwater amount, wherein the estimation formula of [Equation 2] is determined and a rainwater amount is calculated to estimate a three-dimensional distribution of rainfall intensity and rainwater amount . A three-dimensional distribution estimation device for rainfall intensity and rainfall, which estimates the three-dimensional distribution of rainfall intensity and rainfall based on the phase difference K _DP , reflection factor difference Z _DR and reflection factor Z _H obtained by multi-parameter radar There,
A selection means for selecting a calculation mode and a calculation method;
Correction means for correcting rain attenuation;
Temperature profile t (r, θ) in the range direction from the ground temperature t ₀ , radar range r, observation elevation angle θ, and standard atmospheric temperature decrease rate Γ (= 0.065 ° C./m)
A temperature range profile calculation means for calculating
The specific polarization phase difference K _DP , the reflection factor difference Z _DR , the reflection factor Z _H , the elevation angle θ, and the temperature profile t (r, θ) in the range direction are input and based on the calculation mode selected by the selection means. The rain attenuation is corrected by the correction means and based on the selected calculation method.
Either of the above is used as an estimation formula for rainfall intensity,
Is one of the formulas for estimating the amount of rainwater,
Estimation means for estimating a three-dimensional distribution of the rainfall intensity and the amount of rainwater, and the estimation means includes a coefficient and a power index a _i (i = 1, 1) of each estimation formula considering temperature dependence and elevation angle dependence. 2), a _{i ′} (i = 1, 2), b _i (i = 1, 2), b _{i ′} (i = 1, 2), c _i (i = 1, 2, 3), c _{i ′} (I = 1, 2, 3), d _i (i = 1, 2, 3), d _{i ′} (i = 1, 2, 3) Estimating device. The estimation means may be a K _DP method for calculating the R (K _DP ) and M (K _DP ) on condition that the calculation mode is no attenuation correction, or R (Z _H ) or R (K _DP ). , M (Z _H ) or M (K _DP )
3. The apparatus for estimating a three-dimensional distribution of rainfall intensity and amount of rain water according to claim 2, wherein any one of the composite methods A for performing the calculation is selected. The estimation means is a composite method B for calculating R (Z _H ) or R (K _DP ), M (Z _H ) or M (K _DP ) on the condition that the calculation mode is attenuation correction. Calculate R (K _DP , Z _DR ), M (K _DP , Z _DR ) (K _DP , Z _DR ) method, calculate R (Z _H , Z _DR ), M (Z _H , Z _DR ) 3. The apparatus for estimating a three-dimensional distribution of rainfall intensity and amount of rain water according to claim 2, wherein any one of (Z _H , Z _DR ) method is selected.