Radar rainfall deviation correction algorithm based on combination of probability density matching and climate state ratio method

Abstract

The invention discloses a radar rainfall deviation correction algorithm based on combination of probability density matching and a climate state ratio method. The radar rainfall deviation correction algorithm comprises collection and preprocessing of ground station observation data, PDF deviation correction of radar rainfall and climate state deviation correction of radar rainfall. According to the invention, effective utilization of rainfall observation data of a ground meteorological station is realized, deviation correction is carried out on a radar networking estimation rainfall product by using a ground station grid analysis field and a ground-radar climate state proportion analysis field, and the deviation between a radar rainfall estimation value and real rainfall is effectively corrected. According to the method, the discontinuity and systematic deviation of space-time distribution of a fusion product are reduced, the adaptability of the fusion product to complex terrains and rapid evolution precipitation scenes is enhanced, collaborative optimization correction of multi-scale and multi-type deviation is realized, and the reliability and service applicability of a radar precipitation product are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of precipitation bias correction. Specifically, it is a radar precipitation bias correction algorithm based on the combination of probability density matching and climatological ratio method. BACKGROUND

[0002] In the multi-source precipitation fusion real-time analysis business, the quality of radar precipitation data can greatly affect the accuracy of the fusion results. However, due to the limitations of radar detection principles, environmental interference, and terrain shielding factors, there is often a systematic bias between radar precipitation estimates and true precipitation. Especially in complex terrain areas, the radar beam is blocked by mountains, buildings, etc., resulting in incomplete or distorted precipitation echo signals; while in large-scale precipitation systems, differences in radar hardware performance, attenuation effects, and precipitation particle phase state recognition errors, etc., may cause non-independent systematic bias. Existing research has shown that if the radar precipitation data is not effectively corrected for bias, the fused real-time product is prone to significant discontinuity and systematic deviation in the spatial and temporal distribution, affecting short-term forecast, heavy rain warning and forecast model verification business applications, etc.

[0003] The formation mechanism of radar precipitation bias is complex, and its influencing factors can be summarized into two categories: one is dynamic non-independent systematic bias, which is closely related to radar hardware performance, precipitation system evolution characteristics (such as birth and death time, moving speed) and regional station network density differences. For example, the detection range of a single radar is limited, and the spatial matching accuracy of ground observations and radar data is significantly different between dense sites in eastern China and sparse sites in western China; the second is static terrain shielding bias, which is caused by the obstruction of radar beam propagation path, and in long-term climate statistics, it is manifested as systematic underestimation or overestimation in terrain-related areas. In addition, the spatiotemporal distribution characteristics of radar precipitation bias are also affected by the superimposed factors of seasonal variation, precipitation intensity distribution, and sensor upgrade iteration.

[0004] Current correction methods for radar precipitation bias still have limitations. Traditional methods mainly include two categories: one is the statistical correction technique based on real-time ground observations, such as dynamic calibration or regression analysis. Although this method can partially correct the immediate bias between radar and ground observations, it relies too much on the spatial representativeness of real-time station data, and interpolation errors are easy to occur in sparse areas. Moreover, it does not consider the temporal and spatial evolution characteristics of precipitation systems, making it difficult to stably capture the long-term variation of non-independent bias. The second is the correction scheme based on climatological statistics, such as using historical precipitation ratios to build static correction factors. Although this method can alleviate the systematic bias caused by terrain shielding, it cannot effectively handle the dynamic bias of short-term precipitation processes. Moreover, the spatial resolution and timeliness of the climatic factor are insufficient, making it difficult to meet the rapid response needs of extreme weather events. Existing methods independently apply the above two techniques, leading to a fragmentation in the co-processing of dynamic system bias and static terrain bias, which may cause the problem of imbalance in the correction magnitude or a decrease in spatial consistency. Radar precipitation data plays an irreplaceable role in meteorological operations. On the one hand, it provides high temporal and spatial resolution precipitation structure information for multi-source fusion real-time products, supporting fine weather monitoring and disaster warning. On the other hand, as an important input for numerical model assimilation, its quality directly affects the accuracy of precipitation prediction. However, existing correction techniques cannot simultaneously consider the combined effects of dynamic bias and static bias, especially in complex terrain and rapidly evolving precipitation scenarios. The corrected radar data still has significant uncertainty. For example, 202210066596.X discloses a multi-source precipitation data fusion method that uses CMORPH satellite data, radar data, and automatic weather station data for fusion processing. However, it does not consider dynamic terrain adjustment, and the representativeness of a single data source combination is insufficient for special weather (such as short-term severe convective weather) or extreme terrain (such as high-altitude mountainous areas and plateaus), leading to a decrease in fusion accuracy. Moreover, its time resolution is 1 hour, which cannot meet the needs of short-term heavy precipitation. Therefore, there is an urgent need to develop a new fusion correction algorithm that combines the advantages of dynamic probability density matching and climatological ratio adjustment, breaks through the limitations of single methods, and optimizes the correction of multi-scale and multi-type bias, providing technical support for improving the reliability and operational applicability of multi-source precipitation fusion products. SUMMARY

[0005] To this end, the technical problem to be solved by the present application is to provide a radar precipitation bias correction algorithm based on the combination of probability density matching and climatological ratio method, which can effectively solve the problem that existing technologies cannot co-process dynamic and static bias, and the corrected radar data still has significant uncertainty.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] A radar precipitation bias correction algorithm based on the combination of probability density matching and climatological ratio method, comprising:

[0008] S1, collection and preprocessing of ground station observation data to obtain a ground precipitation grid analysis field;

[0009] S2, PDF bias correction of radar precipitation

[0010] S21, consistent matching analysis of ground observation precipitation and radar estimated precipitation;

[0011] S22, constructing a correction model to adjust the spatiotemporal matching window of the ground and radar precipitation PDF samples;

[0012] S23, using the ground precipitation grid analysis field to perform PDF bias correction on the radar network estimated precipitation data;

[0013] S3, climate state bias correction of radar precipitation

[0014] S31, for areas where radar precipitation is severely obstructed by terrain and has poor quality, using the INCA algorithm to correct by calculating the radar precipitation climate proportion correction factor RFC, the calculation formula is:

[0015]

[0016] where P i,j is the ground observation precipitation value at a specified grid point, and P Radar,i,j is the radar estimated precipitation value at a specified grid point;

[0017] S32, constructing a correction model, using the ground-radar climate state proportion analysis field to perform bias correction on the radar network estimated precipitation data, applying the RFC to the original radar estimated precipitation on the grid to obtain the radar estimated precipitation after climate state bias correction, the calculation formula is:

[0018] P * Radar,i,j =RFC i,j *P Radar,i,j (2)。

[0019] Preferably, in the above step S1, the precipitation data of the ground meteorological observation station is collected, quality control processing is performed to obtain quality-controlled ground observation precipitation data processing, and an optimal interpolation method is used to generate a 1km ground grid analysis precipitation product, i.e., a ground precipitation grid analysis field, from the quality-controlled ground observation precipitation data processing.

[0020] Preferably, in the above step S1, the quality-controlled ground observation precipitation data processing method is to use a single data source control module, a multi-source data collaborative quality control module, and a dynamic black list module to perform data quality control.

[0021] Preferably, the single data source control module includes metadata inspection, characteristic value inspection, limit value inspection and dead value inspection, and can eliminate gross errors in real-time precipitation data and dead values that do not change for a long time; the multi-source data collaborative quality control module is a QC algorithm developed based on the consistency of radar, weather phenomenon and other precipitation-related meteorological observation data, and can accurately identify false clear-sky precipitation and false 0-value data in typical precipitation areas that cannot be identified by rapid quality control; after the data processed by the single data source control module and the multi-source data collaborative quality control module, the data are further processed by the dynamic black list module to eliminate observation stations with high error proportion and long error duration, and a flexible dynamic evaluation mechanism is used to remove the data from the black list when the data quality is restored, and select data with quality control codes of 0, 1, 3 and 4 for subsequent fusion analysis.

[0022] Preferably, in the step S1, the 1km ground grid analysis precipitation product is generated by using the optimal interpolation method, including:

[0023] 1) establishing a grid background field of precipitation climate value;

[0024] 2) calculating precipitation ratio data of each station and interpolating to generate a corresponding grid point field, and the ratio data is a new element defined by means of the climate background field: precipitation ratio = observed precipitation of the station / corresponding grid point precipitation climate value;

[0025] 3) multiplying the precipitation ratio grid point field and the corresponding climate background field to generate a precipitation grid point field, wherein the interpolation method for generating the grid point field in step 2) is the optimal interpolation method, and the calculation formula (3) is as follows:

[0026]

[0027] that is, the analysis value A of the grid k is the initial value F of the point k plus the deviation of the observed value and the initial value of the point, and the deviation is obtained by weighted estimation of the deviation of n known initial values F i and observed values O i in the specified analysis range.

[0028] Preferably, in the step S21, the consistency matching analysis includes: (1) data preparation and time alignment, (2) consistency index calculation, (3) optimal lag time determination, and the difference relationship between the 10-minute cumulative precipitation of the ground automatic station observation in the precipitation process and the radar QPE at the lag 0-minute time is analyzed, and the correlation coefficient, root mean square error and relative deviation index are used to determine the optimal lag time.

[0029] Preferably, in the step S22, the space-time matching window parameter is set to 1 hour and 35 km, and the minimum effective sample logarithm participating in the PDF matching is 120.

[0030] Preferably, in the above step S1, the ground precipitation grid analysis field is an equi-latitude-longitude grid interpolated precipitation field generated by using ground station precipitation data.

[0031] Preferably, in the above step S32, the radar network estimated precipitation data is a radar network estimated precipitation product developed by the National Meteorological Information Center by using base data of more than 200 operational weather radars nationwide.

[0032] Preferably, in the above step S32, the ground-radar climatic proportion analysis field is an analysis grid climatic proportion correction factor calculated by using a ratio of cumulative precipitation of recent three months of historical ground observation and radar estimated precipitation.

[0033] The technical scheme of the present application achieves the following beneficial technical effects:

[0034] 1. The present application realizes effective utilization of ground meteorological station precipitation observation data, and effectively corrects the deviation between radar precipitation estimation value and true precipitation by using ground station grid analysis field and ground-radar climatic proportion analysis field to correct the deviation of radar network estimated precipitation product, reduces the discontinuity and systematic deviation of spatio-temporal distribution of the fusion product, enhances the adaptability of the radar precipitation product to complex terrain and rapidly evolving precipitation scene, realizes collaborative optimization correction of multi-scale and multi-type deviation, and significantly improves the reliability and business applicability of the radar precipitation product.

[0035] 2. The present application can help improve the level of meteorological business prediction and service, and the corrected accurate radar precipitation data can be used as an important input for numerical model assimilation, can improve the accuracy of precipitation prediction, and the time resolution can reach 10 minutes and the spatial resolution can reach 1 km; at the same time, the more accurate high spatio-temporal resolution precipitation structure information is provided for multi-source fusion real-time product, the fine weather monitoring is optimized, which helps to timely detect precipitation anomalies and issue disaster warning in advance, provides strong data support for meteorological business such as intelligent grid short-term prediction, heavy rain warning and decision service, and promotes the overall improvement of meteorological business level. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 FIG. 1 is a radar QPE and ground automatic station observation 10-minute cumulative precipitation relationship diagram of different lag time cases of the radar precipitation deviation correction algorithm based on the combination of probability density matching and climatic proportion ratio method according to the present application (1a, correlation coefficient relationship diagram; 1b, root mean square error relationship diagram; 1c, relative deviation relationship diagram); DETAILED DESCRIPTION

[0037] In this embodiment, the radar precipitation deviation correction algorithm is performed by using the ground precipitation grid analysis field, the ground-radar climatic proportion analysis field and the radar network estimated precipitation data, specifically:

[0038] S1, collection and preprocessing of ground station observation data

[0039] S11, collecting precipitation data of ground meteorological observation stations, performing quality control processing, and obtaining quality-controlled ground observation precipitation data;

[0040] The data quality control is performed by using a single data source control module, a multi-source data collaborative quality control module, and a dynamic black list module, wherein,

[0041] The single data source control module includes metadata inspection, characteristic value inspection, limit value inspection, and dead value inspection, and can quickly and efficiently eliminate gross errors in real-time precipitation data and long-time unchanged dead values;

[0042] The multi-source data collaborative quality control module is a QC algorithm developed based on the consistency of radar, weather phenomenon, and various precipitation-related meteorological observation data, and can accurately identify and quickly quality control false clear sky precipitation and false 0 value data in typical precipitation areas;

[0043] The data quality controlled by the single data source control module and the multi-source data collaborative quality control module is further processed by the dynamic black list module to eliminate observation stations with high error proportion and long error duration, and a flexible dynamic evaluation mechanism is adopted, and when the data quality is restored, the data with quality control codes of 0, 1, 3, and 4 (i.e. correct, suspicious, revised, and modified observation data) are selected to participate in subsequent fusion analysis (i.e. the selection strategy of hourly ground precipitation observation data in the ART_1km business system is the same).

[0044] S22, using an optimal interpolation method to process the quality-controlled ground observation precipitation data to generate 1km ground grid analysis precipitation products;

[0045] The climate background field ratio optimal interpolation (OI) gridding analysis technology (Xie et al. 2007; Chen et al. 2002; Shen Yan et al. 2010, Shen Yan et al. 2012) is used to process and generate 1km ground grid analysis precipitation products, including:

[0046] 1) Establishing a grid background field of precipitation climate values;

[0047] 2) Calculating precipitation ratio data of each station and interpolating to generate corresponding grid point fields, and the ratio data is a new element defined by means of the climate background field: precipitation ratio = observed precipitation of the station / corresponding grid point precipitation climate value;

[0048] 3) Multiplying the precipitation ratio grid point field and the corresponding climate background field to generate a precipitation grid point field, wherein the interpolation method applied in step 2) is the optimal interpolation (OI) method, and the calculation formula (3) is as follows:

[0049]

[0050] i.e. the analysis value A of the grid k is the initial value F of the point k plus the deviation of the observation value of the point from the initial value, which is estimated by weighting the deviation of n known initial values F i and observation values O i in the specified analysis range;

[0051] S2, PDF bias correction of radar precipitation

[0052] S21, consistent matching analysis between ground observation precipitation and radar estimated precipitation, including:

[0053] (1) data preparation and time alignment; (2) consistency index calculation; (3) optimal lag time determination. Analyze the difference between the 10-minute cumulative precipitation observed by the ground automatic station every minute during the precipitation process and the radar QPE at the lag 0 minute time, and determine the optimal lag time by using the correlation coefficient, root mean square error, and relative deviation index. This analysis quantifies the consistency of radar and ground observation at different time lags, verifies the reliability of time matching at the 10-minute scale, and provides a prerequisite for the application of the PDF matching method. The core value lies in excluding the interference of time misalignment on bias correction and ensuring that the correction result truly reflects the error characteristics of radar data, rather than the pseudo-bias caused by time sampling difference.

[0054] S22, construct a correction model and adjust the spatiotemporal matching window of ground and radar precipitation PDF samples;

[0055] If the matching window is not set reasonably, the following problems may occur:

[0056] (1) precipitation system birth and death characteristics: short-time heavy precipitation systems (such as thunderstorms) have short life cycles (usually <1 hour). If the matching window is too large (such as 1 hour), it will mix observation data of different precipitation stages, masking the true bias. If the window is too small (such as single minute), the sample size is insufficient, and the statistical reliability is reduced. Therefore, the spatiotemporal matching window parameter is set to 1 hour.

[0057] (2) Radar detection range limit: the effective detection range of a single radar is limited (usually about 200-300 kilometers in diameter), and the precipitation in the edge area may be biased due to beam blocking or distance attenuation, which needs to be adjusted by matching the window to adjust the spatial representativeness. In addition, the difference in station network density: the density of rain gauges in eastern China is high (about 10-20 kilometers apart), and the western part is sparse (> 50 kilometers apart), which needs to balance the spatial coverage and sample representativeness by matching the window to avoid overfitting due to too small window in the east and smoothing distortion due to too large window in the west. Therefore, the spatial window parameter is set to 35 km, and the minimum effective sample number for PDF matching is 120 to ensure statistical reliability and avoid extreme bias caused by small samples. By considering the birth and death time characteristics of the 10-minute precipitation system, the detection range of a single radar, the stability of radar bias variation, and the variation of station network density in different regions of China, the parameters such as the spatiotemporal matching window of ground and radar precipitation PDF samples are adjusted to improve the universality and accuracy of bias correction.

[0058] S23, using ground precipitation grid analysis field to correct the radar network estimated precipitation data PDF bias;

[0059] S3, climatological bias correction of radar precipitation

[0060] S31, for areas where radar precipitation is severely affected by terrain shielding and has poor quality, use INCA algorithm to correct by calculating radar precipitation climatological scaling factor RFC (Climatological Radar scaling Factor, RFC). The analysis grid climatological scaling factor (RFC) is calculated by the ratio of the cumulative precipitation of the latest 3 months of historical ground observation and radar estimated precipitation, and the calculation formula is:

[0061]

[0062] Where P i,j is the ground observation precipitation value at the specified grid point, and P Radar,i,j is the radar estimated precipitation value at the specified grid point.

[0063] S32, construct a correction model and use the ground- radar climatological proportion analysis field to correct the radar network estimated precipitation data bias. Apply RFC to the original radar estimated precipitation of the grid to obtain the radar estimated precipitation after climatological bias correction, and the calculation formula is:

[0064] P * Radar,i,j = RFC i,j *P Radar,i,j (2).

[0065] Among them, the ground precipitation grid analysis field is an interpolated precipitation field with equal latitude and longitude grids generated using precipitation data from ground stations.

[0066] The ground-radar climate ratio analysis field is a climate ratio correction factor for the analysis grid calculated using the ratio of historical ground observations and radar-estimated cumulative precipitation over the past three months.

[0067] The radar network-based precipitation estimation data is a radar network-based precipitation estimation product developed by the National Meteorological Information Center using basic data from more than 200 operational weather radars across the country.

[0068] The present invention is provided in the following information for specific implementation.

[0069] like Figure 1 This study examines the relationship between the cumulative precipitation observed by automatic ground stations over the first 10 minutes and the radar QPE at a lag of 0 minutes during a precipitation event, checking the consistency between radar QPE and ground-based precipitation observations on a 10-minute timescale. The correlation coefficient ( Figure 1 a) Root mean square error ( Figure 1 b) Relative deviation ( Figure 1 c) Indicators show that the cumulative precipitation observed on the ground in the first 10 minutes with a lag of 4 minutes is most consistent with the instantaneous precipitation estimated by the radar at a lag of 0 minutes. Within the lag of 0-9 minutes, the time difference in cumulative time sampling has a relatively small impact on the bias, only 0.2% overall. This indicates that the consistency between radar-detected precipitation and ground observation time matching is good at the 10-minute scale, making it suitable for using the PDF matching method for bias correction. Secondly, considering the characteristics of the formation and dissipation time of precipitation systems every 10 minutes, the detection range of a single radar, the stability of radar bias changes, and the variations in station density in different regions of eastern and western China, the spatiotemporal matching window for ground and radar precipitation PDF samples was adjusted.

[0070] By analyzing the 24-hour cumulative precipitation (unit: mm) observed on July 31, 2021, and comparing it with the 24-hour cumulative precipitation (unit: mm) before and after the bias correction of the radar QPE product PDF on July 31, 2021, it can be seen that, overall, the spatial morphology of precipitation estimated by the radar and the analysis of ground observations are quite similar. However, the radar estimates the strong precipitation centers in Henan, Shandong, and southern Hebei (Hebei-Shandong-Henan region), while the precipitation in Northeast China is underestimated. After bias correction, the intensity of the precipitation center in the Hebei-Shandong-Henan region by the radar is significantly weakened, while the precipitation in Northeast China is slightly strengthened, which is more consistent with ground observations. Statistics are shown in Table 1.

[0071] Table 1. Statistical values ​​of errors in radar and satellite precipitation products before and after correction, taken every 10 minutes as of July 31, 2021.

[0072]

[0073] Statistically, after the bias correction of radar QPE product PDF, CC increases, RMSE decreases, and Bias becomes smaller, which shows that the bias correction can improve the accuracy of radar products to a certain extent.

[0074] Before correction, the original radar precipitation in the mountainous area of the southwest of Sichuan Basin is significantly smaller due to the effect of terrain shielding, and the corresponding RFC value is larger. After correction, the precipitation intensity in this area increases and is close to that in the western part of the basin.

[0075] When the climatological bias correction is selected, sometimes there will be an "over-correction" problem in local areas when compared with the precipitation observations of ground stations and radar QPE products. This is because when the radar shielding is more serious, the too small radar cumulative amount as the denominator causes an abnormally large RFC value. In addition, radar has relatively good detection capability for heavy rain, and we should avoid the weakening of heavy rain by RFC correction. Therefore, the idea of introducing PDF correction is introduced to adjust and limit RFC for strong precipitation (greater than 1 mm / 10 min) and weak precipitation (less than 0.2 mm / 10 min) that reach a certain intensity threshold, to avoid over-correction of radar-detecting heavy rain, while reducing the overestimation of weak precipitation.

[0076] Comparing the effects of different correction methods, the PDF algorithm can better correct the bias of radar in the case of dense station network, but its effect is poor in the complex terrain area of southwest China. The RFC method can effectively correct the bias caused by terrain shielding. However, the RFC method will exacerbate the overestimation of small precipitation, which is easy to cause a positive bias of the overall fused precipitation. In order to take advantage of different correction methods, the bias correction scheme of radar precipitation is adjusted, that is, the RFC method is mainly used in the southwest area to solve the bias caused by terrain shielding, and the PDF method is mainly used in areas with high quality radar to solve the bias caused by different weather and precipitation systems.

[0077] The application utilizes the ground station grid analysis field and the ground-radar climatological proportion analysis field to correct the bias of the radar network estimated precipitation product, and compares and analyzes the feasibility of the PDF bias correction and the effect of different correction methods. From the results of different correction methods, it can be seen that the PDF bias correction and the climatological bias correction can improve the quality of the radar precipitation product to a certain extent, but from the test results, the two methods have certain limitations. Although the PDF algorithm can better correct the bias of the radar in the case of dense station network, the effect is poor in the complex terrain area of southwest China. The RFC method can effectively correct the bias caused by the terrain shielding effect. However, the RFC method will aggravate the overestimation of small precipitation, and easily cause the positive bias of the overall fused precipitation.

[0078] The independent test of the correction results is carried out by retaining more than 2400 national station precipitation observation data. From the statistical analysis results, in the complex terrain area of southwest China and the area with low quality of radar, the comprehensive optimized correction method shows better correction effect. As shown in Table 2, in the complex terrain area of southwest China, the relative deviation of the comprehensive optimized correction result is-1.8%, which is better than that of the PDF bias correction and the climatological bias correction (-18.6% and-3.8%); in the area with low quality of radar, the comprehensive optimized correction result improves the relative deviation (the PDF bias correction and the climatological bias correction are-77.1% and-49.4% respectively) to-46.5%, and improves the relative coefficient (the PDF bias correction and the climatological bias correction are 0.269 and 0.375 respectively) to 0.419. From the spatial distribution of the corrected precipitation and the statistical analysis of the evaluation, the comprehensive optimized correction effect improves the precision of the radar product more, is more consistent with the ground observation, and improves the reliability and business applicability of the radar precipitation product.

[0079] Table 2 Test results of different bias correction schemes in southwest China and areas with poor quality of radar

[0080]

[0081] On the one hand, the application realizes effective utilization of the ground meteorological station precipitation observation data, corrects the bias of the radar network estimated precipitation product by means of various analysis fields such as ground station grid analysis field, corrects the estimation bias, realizes multi-scale and multi-type bias collaborative optimization correction, and improves the product reliability and applicability; on the other hand, the corrected data can improve the accuracy of precipitation forecast, provide accurate information for multi-source fusion real-time product, optimize weather monitoring, provide strong support for many meteorological businesses, and promote the overall improvement of meteorological business.

[0082] Obviously, the above embodiments are only examples for clearly illustrating the present application and are not intended to limit the present application. Based on the above description, one of ordinary skill in the art can make other different forms of changes or modifications. Here, it is not necessary or possible to enumerate all the embodiments. The obvious changes or modifications derived from the above should be covered in the protection scope of the present application.

Claims

1. A radar precipitation bias correction algorithm based on the combination of probability density matching and climatological ratio method, characterized in that, Comprise: S1, collection and preprocessing of ground station observation data to obtain a ground precipitation grid analysis field; S2, PDF bias correction of radar precipitation S21, consistent matching analysis of ground observation precipitation and radar estimated precipitation; S22, construction of a correction model to adjust the temporal and spatial matching window of the ground and radar precipitation PDF samples; S23, PDF bias correction of radar network estimated precipitation data using the ground precipitation grid analysis field; S3, climate bias correction of radar precipitation S31, for areas where radar precipitation is severely obstructed by terrain and the quality is poor, the INCA algorithm is used to calculate the climate proportion correction factor RFC to correct the radar precipitation, and the calculation formula is: where P i,j is the ground-observed precipitation value at the specified grid point location, P Radar,i,j is the radar-estimated precipitation value at the specified grid point location; S32, a correction model is constructed, and the ground-radar climate proportion analysis field is used to correct the bias of the radar network estimated precipitation data, and the RFC is applied to the original radar estimated precipitation of the grid to obtain the radar estimated precipitation after climate bias correction, and the calculation formula is: P * Radar,i,j = RFC i,j *P Radar,i,j (2).

2. The radar precipitation bias correction algorithm based on the combination of the probability density matching and the climatological ratio method according to claim 1, characterized in that, In the step S1, the ground meteorological observation station precipitation data is collected, quality control processing is performed, and the quality controlled ground observation precipitation data processing is generated by using the optimal interpolation method to generate a 1km ground grid analysis precipitation product, i.e., a ground precipitation grid analysis field.

3. The radar precipitation bias correction algorithm based on the combination of the probability density matching and the climatological ratio method according to claim 2, characterized in that, The quality control method of the quality controlled ground observation precipitation data processing is to use a single data source control module, a multi-source data collaborative quality control module and a dynamic black list module to control the data quality.

4. The radar precipitation bias correction algorithm based on the combination of probability density matching and climatological ratio method according to claim 3, characterized in that, The single data source control module includes metadata checking, characteristic value checking, limit value checking and dead value checking, which can eliminate gross errors and long-time unchanged dead values in real-time precipitation data; the multi-source data collaborative quality control module is a QC algorithm developed based on the consistency of radar, weather phenomenon and various precipitation related meteorological observation data, which can accurately identify false clear sky precipitation and false 0 value data in typical precipitation areas which cannot be identified by fast quality control; The data processed by the single data source control module and the multi-source data collaborative quality control module is further processed by the dynamic black list module to eliminate observation stations with high error proportion and long error duration, and a flexible dynamic evaluation mechanism is used to remove the data with quality control codes 0, 1, 3 and 4 from the black list when the data quality is restored.

5. The radar precipitation bias correction algorithm based on the combination of the probability density matching and the climatological ratio method according to claim 3, characterized in that, In the step S1, the optimal interpolation method is used to process and generate a 1km ground grid analysis precipitation product, including: 1) establishing a grid background field of precipitation climate value; 2) calculating the precipitation ratio data of each station and interpolating to generate the corresponding grid point field, and the ratio data is a new element defined by means of the climate background field: precipitation ratio = observed precipitation of the station / corresponding grid precipitation climate value; 3) multiplying the precipitation ratio grid point field and the corresponding climate background field to generate a precipitation grid point field, wherein the interpolation method applied in step 2) is the optimal interpolation method, and the calculation formula (3) is as follows: A is the analysis value of the grid k F is the initial value of the point k The deviation of the observation value and the initial value of the point is added to the initial value F of the point, and the deviation is estimated by weighting the deviation of n known initial values F i and observation values O i in the specified analysis range.

6. The radar precipitation bias correction algorithm based on the combination of probability density matching and climatological ratio method according to claim 1, characterized in that, The consistency matching analysis in the step S21 includes: (1) data preparation and time alignment, (2) consistency index calculation, (3) optimal lag time determination; the difference relationship between the 10-minute cumulative precipitation of the minute-by-minute ground automatic station observation in the precipitation process and the radar QPE at the lag 0-minute time is analyzed, and the optimal lag time is determined by using the correlation coefficient, the root mean square error and the relative deviation index.

7. The radar precipitation bias correction algorithm based on the combination of the probability density matching and the climatological ratio method according to claim 1, characterized in that, In the step S22, the space-time matching window parameters are set as 1 hour and 35 km, and the minimum effective sample pair number participating in the PDF matching is 120.

8. The radar precipitation bias correction algorithm based on the combination of the probability density matching and the climatological ratio method according to claim 1, characterized in that, In the step S1, the ground precipitation grid analysis field is an equi-latitude-longitude grid interpolation precipitation field generated by using the ground station precipitation data.

9. The radar precipitation bias correction algorithm based on the combination of the probability density matching and the climatological ratio method according to claim 1, characterized in that, In the step S32, the radar network estimated precipitation data are radar network estimated precipitation products developed by the National Meteorological Information Center by using the base data of more than 200 weather radars in service nationwide.

10. The radar precipitation bias correction algorithm based on the combination of probability density matching and climatological ratio method according to claim 1, characterized in that, In the step S32, the ground-radar climatic proportion analysis field is an analysis grid climatic proportion correction factor calculated by using the ratio of the cumulative precipitation of the recent 3-month historical ground observation and radar estimated precipitation.

Images