Volume 50,Issue 5,2022 Table of Contents
2022, 50(5):611-622.
DOI: 10.19517/j.1671-6345.20210480
Abstract:
Hangzhou Xiasha SA weather radar adds sophisticated detection technology based on the dual polarization upgrade. In order to further improve the accuracy of radar quantitative precipitation estimation, this paper, referring to the hourly rain gauge data, corrects the radar precipitation estimation algorithm model. It establishes a model based on the real-time QPE rain intensity correction method for minutelevel rain gauge data (QPE-ADJUST method for short) and uses rain gauge data to correct the radar’s QPE data bodybybody scan in real-time. Accumulatively complete 1-hour and 3-hours precipitation estimation products, which improves the radar precipitation estimation accuracy. Through the evaluation of radar products and automatic station data, the precipitation estimation effect of the QPE-ADJUST method is statistically analyzed from three aspects: the precipitation estimation algorithm, the influence of the radar resolution and the influence of the body scan cycle speed. The results show that the QPE-ADJUST method is better in the temporal and spatial distribution of precipitation than other algorithms under the conditions of high radar resolution and fast body scanning period, and the reduced error of radar hourly quantitative precipitation estimation from 50% to 20%, which has high estimation accuracy and stability, and operation application value.
Hangzhou Xiasha SA weather radar adds sophisticated detection technology based on the dual polarization upgrade. In order to further improve the accuracy of radar quantitative precipitation estimation, this paper, referring to the hourly rain gauge data, corrects the radar precipitation estimation algorithm model. It establishes a model based on the real-time QPE rain intensity correction method for minutelevel rain gauge data (QPE-ADJUST method for short) and uses rain gauge data to correct the radar’s QPE data bodybybody scan in real-time. Accumulatively complete 1-hour and 3-hours precipitation estimation products, which improves the radar precipitation estimation accuracy. Through the evaluation of radar products and automatic station data, the precipitation estimation effect of the QPE-ADJUST method is statistically analyzed from three aspects: the precipitation estimation algorithm, the influence of the radar resolution and the influence of the body scan cycle speed. The results show that the QPE-ADJUST method is better in the temporal and spatial distribution of precipitation than other algorithms under the conditions of high radar resolution and fast body scanning period, and the reduced error of radar hourly quantitative precipitation estimation from 50% to 20%, which has high estimation accuracy and stability, and operation application value.
2022, 50(5):623-629.
DOI: 10.19517/j.1671-6345.20210375
Abstract:
The thunderstorm identification algorithm of radar data is an important part of the thunderstorm tracking technique. The traditional continuous region method can only adjust thunderstorm identification results by changing the reflectivity threshold, which cannot meet the current application requirements. This paper proposes a storm identification method based on the OPTICS (Ordering Points to Identify the Clustering Structure) algorithm. This method can identify storm cells based on the density information of points with high reflectivity. Using volume scan data of high-resolution X-band weather radar in two thunderstorms, the performance of the proposed algorithm is tested and compared with the traditional method. The results show that this method can overcome the problems that may occur in traditional methods in high-resolution radar data, such as being unable to distinguish adjacent cells or getting too scattered identification results. Besides, it can flexibly adjust the output results without changing the reflectivity threshold to meet the needs of different applications.
The thunderstorm identification algorithm of radar data is an important part of the thunderstorm tracking technique. The traditional continuous region method can only adjust thunderstorm identification results by changing the reflectivity threshold, which cannot meet the current application requirements. This paper proposes a storm identification method based on the OPTICS (Ordering Points to Identify the Clustering Structure) algorithm. This method can identify storm cells based on the density information of points with high reflectivity. Using volume scan data of high-resolution X-band weather radar in two thunderstorms, the performance of the proposed algorithm is tested and compared with the traditional method. The results show that this method can overcome the problems that may occur in traditional methods in high-resolution radar data, such as being unable to distinguish adjacent cells or getting too scattered identification results. Besides, it can flexibly adjust the output results without changing the reflectivity threshold to meet the needs of different applications.
2022, 50(5):630-635.
DOI: 10.19517/j.1671-6345.20210347
Abstract:
Taking the TITAN system as the basic framework, by introducing big data processing technology, the communication mechanism, data processing and storage pattern are optimized. Based on the support of radar synchronous observation and realtime transmission technology, a radar mosaic system for the ten radars in Liaoning Province and surrounding areas is developed. This radar synchronous observation and mosaic system can realize the streaming transmission of radar data. The distributed radar data processing is realized through the Kafka distributed system, and the stability of radar data storage is improved by using the Cassandra database.
Taking the TITAN system as the basic framework, by introducing big data processing technology, the communication mechanism, data processing and storage pattern are optimized. Based on the support of radar synchronous observation and realtime transmission technology, a radar mosaic system for the ten radars in Liaoning Province and surrounding areas is developed. This radar synchronous observation and mosaic system can realize the streaming transmission of radar data. The distributed radar data processing is realized through the Kafka distributed system, and the stability of radar data storage is improved by using the Cassandra database.
2022, 50(5):636-645.
DOI: 10.19517/j.1671-6345.20210499
Abstract:
Based on the observation data of temperature and geopotential heights at each mandatory level and the forecast field data of ECMWF during radiosonde type change parallel observation in Yunnan, the observation data of new and old radiosondes are comparatively analyzed and evaluated by using BIAS, standard deviations, Root Mean Square Error (RMSE) and related coefficients. The results show that the observation data of the new and old radiosonde are in good agreement. Below 100 hPa, the temperature absolute observation BIASs of new and old radiosondes are less than 1.0 ℃, and geopotential heights are less than 30 gpm. Above 100 hPa, the maximum absolute BIASs of temperature of new and old radiosondes are 3.9 ℃, and geopotential heights are 151.0 gpm. Except for individual levels in the lower and higher levels, the dispersions of observation data of new and old radiosondes are basically the same, or the dispersion of new radiosonde is relatively small. The new radiosonde observation data is more consistent with the model data and is more obvious at the middle and high levels. Compared with the model data, the BIASs of temperature of the new and old radiosondes are concentrated at about ±0.7 ℃ and ±0.9 ℃, the maximum RMSEs of them are about 3.0 ℃ and 4.5 ℃, and the average related coefficients are about 0.78 and 0.73, respectively. The maximum BIASs of geopotential heights of new and old radiosondes are 28.5 gpm and 130.9 gpm, the maximum RMSEs of them are about 87.6 gpm and 136.9 gpm, and the average related coefficients are about 0.86 and 0.78, respectively.
Based on the observation data of temperature and geopotential heights at each mandatory level and the forecast field data of ECMWF during radiosonde type change parallel observation in Yunnan, the observation data of new and old radiosondes are comparatively analyzed and evaluated by using BIAS, standard deviations, Root Mean Square Error (RMSE) and related coefficients. The results show that the observation data of the new and old radiosonde are in good agreement. Below 100 hPa, the temperature absolute observation BIASs of new and old radiosondes are less than 1.0 ℃, and geopotential heights are less than 30 gpm. Above 100 hPa, the maximum absolute BIASs of temperature of new and old radiosondes are 3.9 ℃, and geopotential heights are 151.0 gpm. Except for individual levels in the lower and higher levels, the dispersions of observation data of new and old radiosondes are basically the same, or the dispersion of new radiosonde is relatively small. The new radiosonde observation data is more consistent with the model data and is more obvious at the middle and high levels. Compared with the model data, the BIASs of temperature of the new and old radiosondes are concentrated at about ±0.7 ℃ and ±0.9 ℃, the maximum RMSEs of them are about 3.0 ℃ and 4.5 ℃, and the average related coefficients are about 0.78 and 0.73, respectively. The maximum BIASs of geopotential heights of new and old radiosondes are 28.5 gpm and 130.9 gpm, the maximum RMSEs of them are about 87.6 gpm and 136.9 gpm, and the average related coefficients are about 0.86 and 0.78, respectively.
2022, 50(5):646-652.
DOI: 10.19517/j.1671-6345.20210514
Abstract:
The Meteorological Observation Center of the China Meteorological Administration (CMA) has carried out a vertical comprehensive meteorological observation experiment of megacities. In order to collect and manage a large amount of long-term observation data with new ground-based remote sensing equipment, a megacities observation data system platform has been designed and built. Based on Web technology and the MySQL database, the system has the functions of real-time data transmission, monitoring, storage, analysis, display and sharing. It standardizes the management of new remote sensing observation data and makes product diversification convenient to share. The system improves the data support ability of the new remote sensing equipment and the minute profile data products meet the high timeliness requirements of the short-term weather forecast and nowcasting for real-time data. At present, the system has been popularized and applied in some cities to provide comprehensive platform support for the operational operation of ground-based remote sensing equipment.
The Meteorological Observation Center of the China Meteorological Administration (CMA) has carried out a vertical comprehensive meteorological observation experiment of megacities. In order to collect and manage a large amount of long-term observation data with new ground-based remote sensing equipment, a megacities observation data system platform has been designed and built. Based on Web technology and the MySQL database, the system has the functions of real-time data transmission, monitoring, storage, analysis, display and sharing. It standardizes the management of new remote sensing observation data and makes product diversification convenient to share. The system improves the data support ability of the new remote sensing equipment and the minute profile data products meet the high timeliness requirements of the short-term weather forecast and nowcasting for real-time data. At present, the system has been popularized and applied in some cities to provide comprehensive platform support for the operational operation of ground-based remote sensing equipment.
2022, 50(5):653-659.
DOI: 10.19517/j.1671-6345.20210528
Abstract:
In order to accelerate the intensification and cloud transformation of Fujian Province’s meteorological operation system, it is necessary to rely on the Fujian Meteorological Big Data Cloud Platform (CMADaaS·Fujian) to integrate the system. This paper introduces the general situation, overall structure and function of (CMADaaS·Fujian). The technical scheme and realization method of Fujian Meteorological Integrated Operation Platform integrated into CMADaaS are emphasized. Application integration has been realized and put into operation from six aspects: platform integration into CMADaaS, data source switching to CMADaaS, algorithm integration into the processing line, data products stored into CMADaaS, system monitor information stored in the Real-Time-Monitoring-System (Tianjin), and the front-end page transformation. After integration, the whole process of data aggregation, processing and service of the platform is further optimized and standardized, and the operating efficiency and access speed of the system are significantly improved so as to provide references for other operation systems to integrate into CMADaas.
In order to accelerate the intensification and cloud transformation of Fujian Province’s meteorological operation system, it is necessary to rely on the Fujian Meteorological Big Data Cloud Platform (CMADaaS·Fujian) to integrate the system. This paper introduces the general situation, overall structure and function of (CMADaaS·Fujian). The technical scheme and realization method of Fujian Meteorological Integrated Operation Platform integrated into CMADaaS are emphasized. Application integration has been realized and put into operation from six aspects: platform integration into CMADaaS, data source switching to CMADaaS, algorithm integration into the processing line, data products stored into CMADaaS, system monitor information stored in the Real-Time-Monitoring-System (Tianjin), and the front-end page transformation. After integration, the whole process of data aggregation, processing and service of the platform is further optimized and standardized, and the operating efficiency and access speed of the system are significantly improved so as to provide references for other operation systems to integrate into CMADaas.
2022, 50(5):660-669.
DOI: 10.19517/j.1671-6345.20210478
Abstract:
The purpose of this paper is to analyze whether the FSS(Fraction Skill Score) score results obtained by the two methods are different when the forecast precipitation FSS is calculated for the CMA_MESO model, its horizontal resolution is inconsistent with the observed precipitation, two matching methods are adopted to unify the resolution. In this paper, the forecast data are the 6-hour cumulative precipitation of the 3 km resolution CMA_MESO model, and the observed data is 5 km resolution observed precipitation. The two methods of forecast precipitation matching observed precipitation resolution and observed precipitation matching forecast precipitation resolution are adopted, respectively. Four neighbourhood scales are selected: 5 km, 25 km, 51 km and 105 km; and four precipitation thresholds are selected: 0.1 mm, 4 mm, 13 mm and 25 mm. Two groups of FSS scores with different prediction times are obtained. Through analysis, it is found that there is no significant difference in FSS score between the two groups. The results show that when the horizontal resolution of the prediction precipitation for the CMA_MESO model is inconsistent with the observed precipitation, the predicted precipitation can be matched to the observation grid field, and the observation can also be matched to the model grid field. The two methods do not affect the FSS score results.
The purpose of this paper is to analyze whether the FSS(Fraction Skill Score) score results obtained by the two methods are different when the forecast precipitation FSS is calculated for the CMA_MESO model, its horizontal resolution is inconsistent with the observed precipitation, two matching methods are adopted to unify the resolution. In this paper, the forecast data are the 6-hour cumulative precipitation of the 3 km resolution CMA_MESO model, and the observed data is 5 km resolution observed precipitation. The two methods of forecast precipitation matching observed precipitation resolution and observed precipitation matching forecast precipitation resolution are adopted, respectively. Four neighbourhood scales are selected: 5 km, 25 km, 51 km and 105 km; and four precipitation thresholds are selected: 0.1 mm, 4 mm, 13 mm and 25 mm. Two groups of FSS scores with different prediction times are obtained. Through analysis, it is found that there is no significant difference in FSS score between the two groups. The results show that when the horizontal resolution of the prediction precipitation for the CMA_MESO model is inconsistent with the observed precipitation, the predicted precipitation can be matched to the observation grid field, and the observation can also be matched to the model grid field. The two methods do not affect the FSS score results.
2022, 50(5):670-676.
DOI: 10.19517/j.1671-6345.20210151
Abstract:
The Ensemble-based Reduced Dimension Variational (ERDVar) assimilation method can not only reduce the computational cost without solving the tangential model and adjoint model but also provide the “follow dependent” background error covariance matrix. The NMC (National Meteorogical Center, USA) perturbation method and Regional ERDVar (R-ERDVar) are proposed to resolve the initial perturbation and localization in this article. Finally, ERDVar has been applied to the Global Medium-range Numerical Weather Prediction Model T106L19. The results show that: (1) It is effective to obtain higher accuracy in assimilation using ERDVar, as the information of true innovations is extracted. (2) The NMC initial perturbations reflect the structure of forecast errors and cannot decay easily during forecast subsequently, with at least 10% reduction on forecast errors in ERDVar experiments. (3) Compared with the global ERDVar experiments, there is a 14% reduction for all variable RMSE on average in R-ERDVar experiments, with smaller computational cost. Farther more, the combination use of the R-ERDVar method and NMC perturbation samples can make improvements more stable.
The Ensemble-based Reduced Dimension Variational (ERDVar) assimilation method can not only reduce the computational cost without solving the tangential model and adjoint model but also provide the “follow dependent” background error covariance matrix. The NMC (National Meteorogical Center, USA) perturbation method and Regional ERDVar (R-ERDVar) are proposed to resolve the initial perturbation and localization in this article. Finally, ERDVar has been applied to the Global Medium-range Numerical Weather Prediction Model T106L19. The results show that: (1) It is effective to obtain higher accuracy in assimilation using ERDVar, as the information of true innovations is extracted. (2) The NMC initial perturbations reflect the structure of forecast errors and cannot decay easily during forecast subsequently, with at least 10% reduction on forecast errors in ERDVar experiments. (3) Compared with the global ERDVar experiments, there is a 14% reduction for all variable RMSE on average in R-ERDVar experiments, with smaller computational cost. Farther more, the combination use of the R-ERDVar method and NMC perturbation samples can make improvements more stable.
2022, 50(5):677-685.
DOI: 10.19517/j.1671-6345.20210434
Abstract:
In order to improve the accuracy of grid precipitation forecast, aiming at the phenomenon that there are still much empty precipitation forecasts in the numerical model grid precipitation forecast, based on the live grid precipitation, applying the dynamic modelling and machine training, with the monotonicity of normalized new test parameters, and the computational convenience of the new precipitation TS scoring formula, the dynamic optimal elimination of grid-by-grid precipitation is researched. The research show that the empty precipitation threshold is suppressed in the two-step method, and the process of threshold training selection is more direct in the normalized method. With these two methods, the rain or shine accuracy is all increased. Among them, ECMWF (European Centre for Medium-Range Weather Forecasts) increases by 2.39%-4.76%; the TS score of ECMWF improves the most, increasing 2.98%-3.64% during the day and increasing 1.61%-1.78% at night. However, the TS scores in CMA-SH9 and CMA-BJ decline. The normalized method increases the clear/rain forecast accuracy the most during the day. The analysis shows that the drop in the false alarm rate is significantly greater than the missing forecast rate. As a result, the FAR drops significantly, and the clear/rain forecast accuracy also increases significantly.
In order to improve the accuracy of grid precipitation forecast, aiming at the phenomenon that there are still much empty precipitation forecasts in the numerical model grid precipitation forecast, based on the live grid precipitation, applying the dynamic modelling and machine training, with the monotonicity of normalized new test parameters, and the computational convenience of the new precipitation TS scoring formula, the dynamic optimal elimination of grid-by-grid precipitation is researched. The research show that the empty precipitation threshold is suppressed in the two-step method, and the process of threshold training selection is more direct in the normalized method. With these two methods, the rain or shine accuracy is all increased. Among them, ECMWF (European Centre for Medium-Range Weather Forecasts) increases by 2.39%-4.76%; the TS score of ECMWF improves the most, increasing 2.98%-3.64% during the day and increasing 1.61%-1.78% at night. However, the TS scores in CMA-SH9 and CMA-BJ decline. The normalized method increases the clear/rain forecast accuracy the most during the day. The analysis shows that the drop in the false alarm rate is significantly greater than the missing forecast rate. As a result, the FAR drops significantly, and the clear/rain forecast accuracy also increases significantly.
2022, 50(5):686-693.
DOI: 10.19517/j.1671-6345.20210466
Abstract:
Based on data from 84 observation stations in Guizhou Province and NCEP/NCAR reanalysis, a comprehensive comparative analysis of two extreme freezing rains that occurred in 2008 and 2011 are made through temperature declining amplitudes, affected stations, sea surface temperature patterns and circulation patterns. The results show that: the Eastern type of La Nina with medium intensity provided a favourable condition for the development of freezing rain in two cases. Besides, there were some similar circulation patterns in two cases, such as positive geopotential height anomalies over West Asia and negative geopotential height over East Asia at 500 hPa, stable maintenance of shear line at 850 hPa, southwest jet at 700 hPa, temperature inversion layer and melting layer between 800 to 600 hPa, which were conducive to the formation of freezing rain. However, compared to the case in 2011, it could be observed that the shear line located further north, southwest jet at 850 hPa and temperature inversion layer had stronger intensity and larger area, colder temperature advection and thicker melting layer with longer duration and higher centre temperature from the case in 2008, which could possibly result in more significant influence from freezing rains in 2008.
Based on data from 84 observation stations in Guizhou Province and NCEP/NCAR reanalysis, a comprehensive comparative analysis of two extreme freezing rains that occurred in 2008 and 2011 are made through temperature declining amplitudes, affected stations, sea surface temperature patterns and circulation patterns. The results show that: the Eastern type of La Nina with medium intensity provided a favourable condition for the development of freezing rain in two cases. Besides, there were some similar circulation patterns in two cases, such as positive geopotential height anomalies over West Asia and negative geopotential height over East Asia at 500 hPa, stable maintenance of shear line at 850 hPa, southwest jet at 700 hPa, temperature inversion layer and melting layer between 800 to 600 hPa, which were conducive to the formation of freezing rain. However, compared to the case in 2011, it could be observed that the shear line located further north, southwest jet at 850 hPa and temperature inversion layer had stronger intensity and larger area, colder temperature advection and thicker melting layer with longer duration and higher centre temperature from the case in 2008, which could possibly result in more significant influence from freezing rains in 2008.
WU Yingxu
,
ZHANG Libao
,
ZHANG Yingxin
,
ZHOU Yi
,
ZHAO Ning
,
MENG Yingying
,
LUAN Chen
,
ZHAO Guangna
2022, 50(5):694-701.
DOI: 10.19517/j.1671-6345.20210249
Abstract:
An analysis of a continuous and extremely intense precipitation process is made by conventional meteorological observation data, regional automatic station encrypted observation data, ECMWF thin, GRAPES_MESO, FY-4 satellite cloud image, new generation weather radar and NECP 1°×1° reanalysis data in the central and southwestern parts of Heilongjiang from 08:00 on August 6 to 08:00 on August 8, 2019. This paper analyses the dynamic mechanism of the heavy precipitation process and the nature and distribution characteristics of the precipitation caused by it. The results show that: The heavy precipitation process can be divided into three stages and two types of precipitation echoes: convective precipitation in the saddle field connected to the cold vortex; mixed and convective cloud precipitation combined by saddle field and enhanced warm front; convective cloud precipitation formed by changing the moving path of cloud systems under the action of the residual vortex of the typhoon. The interaction of cold vortex, subtropical high pressure and typhoon were the fundamental causes of this process. The subtropical high pressure and the warm and humid airflow outside the typhoon cooperate with the cold vortex and cold air provide water vapour and unstable conditions for heavy precipitation. The narrow water vapour transport channel caused the spatial discontinuity of heavy precipitation. The low-level convergence line provided trigger conditions for heavy precipitation. The stable structure of the saddle field, the forced uplift of the southern foot of the Daxing’an Mountains, and the blockade of typhoons prolonged the duration of heavy rainfall.
An analysis of a continuous and extremely intense precipitation process is made by conventional meteorological observation data, regional automatic station encrypted observation data, ECMWF thin, GRAPES_MESO, FY-4 satellite cloud image, new generation weather radar and NECP 1°×1° reanalysis data in the central and southwestern parts of Heilongjiang from 08:00 on August 6 to 08:00 on August 8, 2019. This paper analyses the dynamic mechanism of the heavy precipitation process and the nature and distribution characteristics of the precipitation caused by it. The results show that: The heavy precipitation process can be divided into three stages and two types of precipitation echoes: convective precipitation in the saddle field connected to the cold vortex; mixed and convective cloud precipitation combined by saddle field and enhanced warm front; convective cloud precipitation formed by changing the moving path of cloud systems under the action of the residual vortex of the typhoon. The interaction of cold vortex, subtropical high pressure and typhoon were the fundamental causes of this process. The subtropical high pressure and the warm and humid airflow outside the typhoon cooperate with the cold vortex and cold air provide water vapour and unstable conditions for heavy precipitation. The narrow water vapour transport channel caused the spatial discontinuity of heavy precipitation. The low-level convergence line provided trigger conditions for heavy precipitation. The stable structure of the saddle field, the forced uplift of the southern foot of the Daxing’an Mountains, and the blockade of typhoons prolonged the duration of heavy rainfall.
12 Comparative Analysis of Two Regional Heavy Rain Formation Mechanisms in Southwest Shandong in 2020
2022, 50(5):702-712.
DOI: 10.19517/j.1671-6345.20210382
Abstract:
Making use of conventional ground observation, upper air observation and ERA5 reanalysis data, the regional heavy rain and its associated short-time heavy rainfall processes in Southwest Shandong Province on July 22 (referred to as “7·22” process) and August 6-7 (referred to as “8·6” process) in 2020 are diagnostically analyzed. The results show that: the “7·22” process was a surface cyclone precipitation process, and heavy rainstorms mainly occurred from the cyclone’s centre to the inverted trough moving to the front right part. The short-time heavy precipitation was caused by the enhancement of inertia instability after the convective instability was triggered. The “8·6” process was a precipitation process in the warm area of the WPSH edge, and heavy rainstorms mainly occurred in the front of the low-level jet and near the surface convergence line. The short-time heavy precipitation was caused by the triggering and release of convective instability. In the “7·22” process, the warm-wet jet was stronger, and the divergence of water vapour flux and dynamic conditions were significantly stronger than those in the process of “8·6”. The strong convergence zone at very low levels, the greater value area of vapor flux divergence, the strong frontogenesis area on the edge of greater value area of horizontal kinetic energy, and the small value area of |MPV2| on the edge of great value area of MPV are very indicative for the emergence of short-time strong rainfall. Both processes show that the combination of vertical upward movement and deep wet area is a good indicator of short-time heavy precipitation occurrence time.
Making use of conventional ground observation, upper air observation and ERA5 reanalysis data, the regional heavy rain and its associated short-time heavy rainfall processes in Southwest Shandong Province on July 22 (referred to as “7·22” process) and August 6-7 (referred to as “8·6” process) in 2020 are diagnostically analyzed. The results show that: the “7·22” process was a surface cyclone precipitation process, and heavy rainstorms mainly occurred from the cyclone’s centre to the inverted trough moving to the front right part. The short-time heavy precipitation was caused by the enhancement of inertia instability after the convective instability was triggered. The “8·6” process was a precipitation process in the warm area of the WPSH edge, and heavy rainstorms mainly occurred in the front of the low-level jet and near the surface convergence line. The short-time heavy precipitation was caused by the triggering and release of convective instability. In the “7·22” process, the warm-wet jet was stronger, and the divergence of water vapour flux and dynamic conditions were significantly stronger than those in the process of “8·6”. The strong convergence zone at very low levels, the greater value area of vapor flux divergence, the strong frontogenesis area on the edge of greater value area of horizontal kinetic energy, and the small value area of |MPV2| on the edge of great value area of MPV are very indicative for the emergence of short-time strong rainfall. Both processes show that the combination of vertical upward movement and deep wet area is a good indicator of short-time heavy precipitation occurrence time.
2022, 50(5):713-723.
DOI: 10.19517/j.1671-6345.20210476
Abstract:
On the night of June 9th and June 12th, 2021, Shaoxing, Zhejiang Province, experienced two warm-sector heavy rainfall. Both subjective and objective forecasts show large deviations. Using the ground-based automatic station, Doppler radar observation data and ERA5 reanalysis data, the circulation field, triggering mechanism and mesoscale convective system evolution of the two processes are analyzed. The results are as follows: (1) Both processes occurred under the background of weak weather forcing. The “609” process took place under the action of boundary layer jets. While the “612” process occur red at the edge of the subtropical high. (2) During the “609” process, the centre of the large vertical helicity stretched up to the middle troposphere, and the upstream potential vorticity disturbance was continuous transportation, promoting the continuous development of the potential vortex disturbance in the rainstorm area. The large vertical helicity centre of the “612” process only extended to the lower troposphere, and there was dry and cold air intrusion on the isentropic surface, which was conducive to the rapid strengthening of the potential vortex disturbance. (3) The triggering mechanisms of the two processes were both β mesoscale shear lines. The mesoscale convective system developed continuously along the shear line and was in a “quasi-static” state, causing heavy rainstorms. The β mesoscale shear line and the mesoscale convective system of the “609” process had a longer stagnation time than the “612” process and were affected by the windward slope’s uplift, so the total amount of precipitation was greater.
On the night of June 9th and June 12th, 2021, Shaoxing, Zhejiang Province, experienced two warm-sector heavy rainfall. Both subjective and objective forecasts show large deviations. Using the ground-based automatic station, Doppler radar observation data and ERA5 reanalysis data, the circulation field, triggering mechanism and mesoscale convective system evolution of the two processes are analyzed. The results are as follows: (1) Both processes occurred under the background of weak weather forcing. The “609” process took place under the action of boundary layer jets. While the “612” process occur red at the edge of the subtropical high. (2) During the “609” process, the centre of the large vertical helicity stretched up to the middle troposphere, and the upstream potential vorticity disturbance was continuous transportation, promoting the continuous development of the potential vortex disturbance in the rainstorm area. The large vertical helicity centre of the “612” process only extended to the lower troposphere, and there was dry and cold air intrusion on the isentropic surface, which was conducive to the rapid strengthening of the potential vortex disturbance. (3) The triggering mechanisms of the two processes were both β mesoscale shear lines. The mesoscale convective system developed continuously along the shear line and was in a “quasi-static” state, causing heavy rainstorms. The β mesoscale shear line and the mesoscale convective system of the “609” process had a longer stagnation time than the “612” process and were affected by the windward slope’s uplift, so the total amount of precipitation was greater.
2022, 50(5):724-733.
DOI: 10.19517/j.1671-6345.20210455
Abstract:
In this paper, the convolutional neural network and gated recurrent units neural network are used to conduct lightning forecasting research based on radar reflectivity factors and lightning location data. First, a deep learning model (Attention-ConvGRU) based on the convolutional neural network and gated recurrent unit neural network that introduces the attention mechanism is constructed. Then, the radar reflectivity factor data and the lightning location data of the corresponding period (6 minutes) are processed into image data, and input into the deep learning model to train the models that can predict lightning, including three models: single lightning data model, single radar data model and lightningradar dual data model. Finally, forecasting experiment and quantitative evaluation are carried out. The comprehensive evaluation shows that the forecasting model has a comprehensive forecasting accuracy of 96.74%, a false alarm rate of 35.83%, and a Critical Success Index (CSI) of 0.2072. The case study shows that the forecasting model has better lightning forecasting skills for thunderstorms with obvious moving trends (type A thunderstorms) than those without obvious moving trends (type B thunderstorms), and the forecasting skill of the model gradually weakens as the intensity of type B thunderstorms weakens.
In this paper, the convolutional neural network and gated recurrent units neural network are used to conduct lightning forecasting research based on radar reflectivity factors and lightning location data. First, a deep learning model (Attention-ConvGRU) based on the convolutional neural network and gated recurrent unit neural network that introduces the attention mechanism is constructed. Then, the radar reflectivity factor data and the lightning location data of the corresponding period (6 minutes) are processed into image data, and input into the deep learning model to train the models that can predict lightning, including three models: single lightning data model, single radar data model and lightningradar dual data model. Finally, forecasting experiment and quantitative evaluation are carried out. The comprehensive evaluation shows that the forecasting model has a comprehensive forecasting accuracy of 96.74%, a false alarm rate of 35.83%, and a Critical Success Index (CSI) of 0.2072. The case study shows that the forecasting model has better lightning forecasting skills for thunderstorms with obvious moving trends (type A thunderstorms) than those without obvious moving trends (type B thunderstorms), and the forecasting skill of the model gradually weakens as the intensity of type B thunderstorms weakens.
2022, 50(5):734-741.
DOI: 10.19517/j.1671-6345.20210361
Abstract:
The relationship between topographic information data (land cover type, elevation, aspect and slope) and the parameters such as the frequency of CG (Cloud-to-Ground) lightning, the positive ratio of CG lightning, lightning return stroke intensity and lightning return stroke steepness is quantitatively studied based on the CG lightning data from 2007 to 2019, land cover type data and the digital elevation model data. The results show that CG lightning frequency per unit area and the proportion of small-amplitude CG lightning in the built-up area are relatively high, which are 87.1 times/km2 and 8.6%, respectively. The CG lightning return stroke steepness in cultivated areas and water areas is higher than those in other land cover types. The return stroke of CG lightning in Hubei Province is mainly concentrated at an altitude of 0 to 700m and a slope interval of 0.5° to 35°. The return stroke density of CG lightning generally decreases with the altitude increase. The average return stroke intensity of CG lightning first increases and then decreases with altitude increase. The average return stroke steepness of CG lightning decreases with the altitude increase. The CG lightning return stroke density is higher in the south or east of the slope. CG lightning’s average return stroke intensity is higher in the south or north of the slope. The lightning parameter changes in different slope and altitude intervals are relatively consistent.
The relationship between topographic information data (land cover type, elevation, aspect and slope) and the parameters such as the frequency of CG (Cloud-to-Ground) lightning, the positive ratio of CG lightning, lightning return stroke intensity and lightning return stroke steepness is quantitatively studied based on the CG lightning data from 2007 to 2019, land cover type data and the digital elevation model data. The results show that CG lightning frequency per unit area and the proportion of small-amplitude CG lightning in the built-up area are relatively high, which are 87.1 times/km2 and 8.6%, respectively. The CG lightning return stroke steepness in cultivated areas and water areas is higher than those in other land cover types. The return stroke of CG lightning in Hubei Province is mainly concentrated at an altitude of 0 to 700m and a slope interval of 0.5° to 35°. The return stroke density of CG lightning generally decreases with the altitude increase. The average return stroke intensity of CG lightning first increases and then decreases with altitude increase. The average return stroke steepness of CG lightning decreases with the altitude increase. The CG lightning return stroke density is higher in the south or east of the slope. CG lightning’s average return stroke intensity is higher in the south or north of the slope. The lightning parameter changes in different slope and altitude intervals are relatively consistent.
2022, 50(5):742-750.
DOI: 10.19517/j.1671-6345.20210513
Abstract:
In order to reinforce the scientific part of the prevention and mitigation of rainstorm disasters, based on the hourly precipitation data of 126 national weather stations from 2010 to 2019 and basic geographic information, we build the model of risk assessment of rainstorm disasters through three aspects: the danger factor of disaster-inducing, the sensitivity of disaster-pregnant environment and the vulnerability of the disaster-bearing body. To realize the division of risk areas relating to rainstorm disasters in Yunnan, we apply four methods: maximum entropy, natural break point, ArcGis interpolation and grid analysis. Results show that the high-risk areas of rainstorm disasters centre in the southern Yunnan, including Xishuangbanna, Pu’er, the southern Honghe and Dehong. However, Diqing, Nujiang and the northern Lijiang are lower-risk areas. Throughout the whole rainstorm disaster in Yunnan, high-risk and secondary high-risk areas account for 7.05% and 25.22%, and low-risk and secondary low-risk areas account for 10.32% and 21.86%. Using the data, such as the times of rainstorm disasters in 2020 and the testing results of area-division of rainstorm disasters, it is shown that the area-division assessment is scientific and rational.
In order to reinforce the scientific part of the prevention and mitigation of rainstorm disasters, based on the hourly precipitation data of 126 national weather stations from 2010 to 2019 and basic geographic information, we build the model of risk assessment of rainstorm disasters through three aspects: the danger factor of disaster-inducing, the sensitivity of disaster-pregnant environment and the vulnerability of the disaster-bearing body. To realize the division of risk areas relating to rainstorm disasters in Yunnan, we apply four methods: maximum entropy, natural break point, ArcGis interpolation and grid analysis. Results show that the high-risk areas of rainstorm disasters centre in the southern Yunnan, including Xishuangbanna, Pu’er, the southern Honghe and Dehong. However, Diqing, Nujiang and the northern Lijiang are lower-risk areas. Throughout the whole rainstorm disaster in Yunnan, high-risk and secondary high-risk areas account for 7.05% and 25.22%, and low-risk and secondary low-risk areas account for 10.32% and 21.86%. Using the data, such as the times of rainstorm disasters in 2020 and the testing results of area-division of rainstorm disasters, it is shown that the area-division assessment is scientific and rational.
External LinksChina Meteorological AdministrationCMA Meteorological Observation CentreChinese Academy of Meteorological SciencesBeijing Meteorological ServiceNational Satellite Meteorological Center
You are thevisitors
Supported by:Beijing E-Tiller Technology Development Co., Ltd.
Supported by:Beijing E-Tiller Technology Development Co., Ltd.