Processing math: 100%

Evaluation of High-Resolution FY-4B AMV and Temperature Products in Tracking the Northeast China Cold Vortex

FY-4B卫星高分辨率云导风和温度产品识别东北冷涡

+ Author Affiliations + Find other works by these authors
Funds: 
Supported by the China Meteorological Administration Northeast China Cold Vortex Research Key Laboratory (2023SYIAEKFZD04) and Research Project of China Meteorological Administration Training Centre (2024CMATCQN03 and 2024CMATCPY01).

PDF

  • The Northeast China Cold Vortex (NECV) is a significant atmospheric circulation system that triggers severe weather in mid-to-high latitudes of Asia. Fengyun-4B (FY-4B) satellite provides 15-min atmospheric motion vector (AMV) and 2-h three-dimensional temperature profiles, enabling unprecedented high spatiotemporal resolution for real-time vortex tracking. This study evaluates the effectiveness of FY-4B AMV and temperature products in tracking 24 NECVs in 2023, among which two strong NECVs in winter and summer 2023 were carefully examined. We first assessed the accuracy of wind speed and direction of the AMVs in the NECV monitoring region by comparing them with radiosonde observations, revealing reasonable correlation coefficients (CC), mean absolute errors (MAE), and root mean square errors (RMSE). NECVs and their centers were identified by using AMV data from four channels (CH09, CH10, CH11, and CH13) within the 200–500-hPa layer, employing the “8-point method” that sets specific criteria for the wind directions at 8 surrounding points to ensure a consistent cyclonic pattern around the central point. The NECV centers identified from AMVs are found to be close (mean distance of 181.9 km) to those determined by ERA5 geopotential height. The retrieved FY-4B temperature data are also evaluated against radiosonde observations, showing a high CC of 0.996 and RMSE of 1.87 K, indicating reliable temperature retrievals for NECV tracking. Based on the FY-4B/Geostationary Interferometric Infrared Sounder (GIIRS) 500-hPa temperature, the NECV cold centers are obtained and cross-validated against ERA5 reanalysis temperature at 500 hPa, revealing a mean distance deviation of 140.6 km. The real-time operational NECV monitoring based on the FY-4B AMV and temperature products on high spatiotemporal resolutions in this study provides valuable information for disaster prevention and mitigation.

    东北冷涡(NECV)是亚洲中高纬度地区引发恶劣天气的重要大气环流系统。风云四号B星(FY-4B)提供的15分钟频次云导风(AMV)和2小时频次三维温度廓线,为NECV涡旋实时识别追踪提供了前所未有的高时空分辨率观测数据。本研究评估了FY-4B卫星的AMV和温度产品对2023年24个NECV过程识别追踪的效果,并详细分析了冬季和夏季两个强NECV过程。首先通过与探空观测数据对比,评估了NECV监测区域内CH10通道AMV的风速和风向的准确性,结果显示出合理的相关系数(CC)、平均绝对误差(MAE)和均方根误差(RMSE)。选用200–500 hPa气压层内四个通道(CH10、CH09、CH13和CH11)的AMV数据,采用“8点法”确定NECV中心,该方法为8个邻近点的风向设定了具体标准,以确保中心点周围为气旋式环流。利用AMV确定的NECV中心与由ERA5位势高度确定的NECV中心平均距离为181.9公里。另外,基于探空温度对FY-4B温度进行评估,结果显示CC高达0.996,RMSE为1.87 K,表明利用温度来识别追踪NECV的可靠性。基于FY-4B/GIIRS与ERA5 500 hPa温度识别的NECV冷中心平均距离偏差为140.6 km。本研究基于FY-4B卫星高时空分辨率AMV和温度产品开展NECV实时监测追踪,为防灾减灾提供了有价值的信息。离偏差为140.6 km。本研究基于FY-4B卫星高时空分辨率AMV和温度产品开展NECV实时监测追踪,为防灾减灾提供了有价值的信息。

  • The Northeast China Cold Vortex (NECV) is a synoptic-scale cyclonic cold vortex occurring or moving into the region (35°–60°N, 105°–145°E) in the mid–upper troposphere of the mid-to-high latitude Asia (He et al., 2015). The common life cycle of NECV is 5–7 days. It is one of the most significant triggers of climate anomalies in East Asia and exhibits a distinct “climate effect” (He et al., 2006; Miao et al., 2006a, b). The NECV activities often induce severe convective weather such as hail, strong winds, tornadoes, and localized heavy rain in Northeast, North, central, and South China (Zhang et al., 2019; Zhang et al., 2021). The NECV is also associated with extreme cold events (Yin et al., 2013). Previous studies focus mainly on subjective and objective NECV identification, the NECV structural characteristics, and the mechanisms by which the NECV influences heavy rain and severe convective weather. Investigations of NECVs using satellite data are gradually increasing.

    Subjective and objective approaches are used to identifying NECVs. Subjective identification involves manual recognition using upper-troposphere weather maps (Price and Vaughan, 1992; Sun et al., 1994; Wu et al., 2009), while objective identification is based on data such as geopotential height, temperature, and wind fields on isobaric surfaces, using certain computer algorithms for the automatic identification of the NECV. Zhang (2001) and Zhang and Liu (2010) used the “8-point method” and “three-circle method”, respectively, based on geopotential height to identify the center of the cold vortex. Jiang et al. (2012) and Wang et al. (2012) used multiple parameters such as geopotential height, wind fields, and temperature for identification, considering the closed low-pressure circulation and cold characteristics of the NECV. Huang and Li (2020) combined the subjective analysis of weather maps to objective identification of the NECV, effectively improving the accuracy of NECV automatic identification.

    A well-developed NECV is a deep cyclonic vortex, characterized by cyclonic inflow at lower levels, a cold-core structure throughout the troposphere, and a warm-core structure in the lower stratosphere. The cold center is located in the mid–upper troposphere (Wang et al., 2017). With the enhanced capabilities of Chinese Fengyun meteorological satellites in comprehensive observations, there is great potential for the use of Fengyun satellite data to study the NECV. Satellite data have a wide coverage and, compared to conventional observation data, can offer a good spatial resolution (Ren et al., 2019). High-frequency satellite cloud images can better monitor the structural characteristics of NECV during the vortex activities, and satellite-derived atmospheric motion vectors (AMVs) can quantitatively identify these vortex characteristics (Lin et al., 2023; Ren et al., 2023).

    In many studies, application of satellite data to NECV identification mainly involves using infrared and water vapor images, cloud top brightness temperature, and cloud optical thickness to reveal the structural characteristics of the NECV (Wu et al., 2010; Wang et al., 2015; Shi et al., 2022). Niu et al. (2021) established a network observation system based on FY-3D, NOAA-15, Suomi-NPP, and MetOp-B microwave radiometers to frequently retrieve three-dimensional atmospheric temperature fields, geopotential height fields, and instability indices, thus monitoring the generation and development of the NECV in real time.

    Prior research on cold vortices (e.g., Nieto et al., 2005) primarily focused on statistical climatology and synoptic-scale dynamics using reanalysis or radiosonde data. Satellite-based studies, such as those using GOES or Himawari AMVs, emphasized tropical cyclones or midlatitude systems but rarely addressed NECV-specific tracking (Price and Vaughan, 1992; Durre et al., 2018). Infrared hyperspectral sounders (e.g., AIRS, IASI) have been used for atmospheric profiling, but their application to cold vortex center identification remains limited. NECV research in China has traditionally relied on radiosonde observations, ERA5 reanalysis, or polar-orbiting satellite data (e.g., FY-3D; Niu et al., 2021). Studies conducted by Zhang (2001), Wang et al. (2012), and Huang and Li (2020) have established objective identification methods, such as geopotential height thresholds and wind field criteria; however, these studies did not integrate high-resolution geostationary satellite products into their analyses. Previous evaluations of FY-4A/Geostationary Interferometric Infrared Sounder (GIIRS) temperature (Ren et al., 2022) focused on general accuracy rather than vortex-specific applications.

    The main meteorological parameters for monitoring and identifying the NECV include the geopotential height, wind fields, and temperature. Few previous studies have used satellite retrieved high-resolution AMV and temperature products to track the NECV. This study pioneers the application of China’s new-generation geostationary satellite (FY-4B) to NECV monitoring. Unlike prior work relying on polar-orbiting satellites or reanalysis, FY-4B provides 15-min AMV and 2-h 3D temperature profiles, enabling unprecedented spatiotemporal resolution for real-time vortex tracking.

    This paper describes the application of the AMV and temperature profile products derived from Fengyun geostationary meteorological satellites in identifying the NECV. It is structured as follows. Section 2 describes the data, methods and NCEV center tracking. Section 3 presents the main results regarding the evaluation of satellite-derived products in NECV activity region, selection of pressure layers for FY-4B AMV, and identify the NECV center using satellite products. The final conclusions and discussion are given in Section 4.

    FY-4B satellite, launched in June 2021, is the first operational satellite from the new generation of Chinese geostationary meteorological satellites. FY-4B is equipped with four instruments, three of which are related to weather applications: the Advanced Geostationary Radiation Imager (AGRI), the GIIRS, and the Geostationary High-Speed Imager. In December 2022, the FY-4B satellite officially began operational service in orbit above the equator at 133°E. In March 2024, the FY-4B satellite drifted from 133°E to 105°E, replacing the FY-4A satellite.

    The data used in this paper include AMVs derived from FY-4B/AGRI and temperature profile products derived from FY-4B/GIIRS. The data cover the year 2023 and can be downloaded at http://data.nsmc.org.cn. This study combines 500-hPa temperature minima (cold-core identification) with AMV-derived cyclonic circulation, addressing limitations of purely geopotential height-based approaches.

    (1) FY-4B/AGRI AMV

    Unlike earlier studies using single-channel wind retrievals, this work prioritizes AMV data from four channels (CH09, CH10, CH11, and CH13) based on vertical accuracy (200–500 hPa), optimizing vortex center detection via the 8-point method.

    The AMV product includes data from four channels: three water vapor channels (CH09, CH10, and CH11, with central wavelengths of 6.25, 6.95, and 7.42 μm, respectively) and one long-wave infrared channel (CH13, with central wavelength of 10.8 μm). The data frequency is 15 min, starting from 0000 (UTC), and covers the entire FY-4B full disk. The horizontal spatial resolution is 64 km. The data are in gridded format, with each record including wind speed, wind direction, and atmospheric pressure. Figure 1 shows the distribution of FY-4B AMVs at 0600 UTC 7 July 2023. The water vapor weighting heights for CH09, CH10, and CH11 decrease sequentially, so the AMVs derived from the CH09 water vapor channel mainly represents the mid-upper tropospheric wind field at 100–399 hPa (red barbs in Fig. 1). The AMVs derived from the CH10 and CH11 water vapor channels cover some of the mid–lower tropospheric wind field at 400–699 hPa (green barbs in Fig. 1). The height of the AMVs derived from the CH13 longwave infrared channel corresponds to the cloud top height. Therefore, compared to water vapor channels, the AMVs from CH13 can obtain more of the mid–lower tropospheric wind field data at 400–950 hPa (green and blue barbs in Fig. 1). This study comprehensively utilizes AMV data from the four channels.

    Fig  1.  FY-4B AMVs from (a) CH09, (b) CH10, (c) CH11, and (d) CH13 at 0600 UTC 7 July 2023.

    (2) FY-4B/GIIRS temperature profile

    The FY-4B/GIIRS is a successor to the FY-4A/GIIRS instrument (Ren et al., 2022). It is the first advanced remote sensing instrument on a geostationary orbit that uses infrared hyperspectral interferometry to observe the three-dimensional vertical structure of the atmosphere. This represents a breakthrough, transitioning from two-dimensional to three-dimensional comprehensive observations on a geostationary orbit. The instrument operates in the longwave infrared band (680–1130 cm−1) and mid-wave infrared band (1650–2250 cm−1) with a spectral resolution of 0.625 cm−1 and a radiometric calibration accuracy of 0.7 K. The observation area covers China and its surrounding regions, with a temporal resolution of 2 h, starting at 0100 UTC, and a sub-satellite point resolution of 12 km. The temperature profile has 101 pressure levels vertically. When identifying the cold center position of the NECV using FY-4B/GIIRS temperature, only data with quality flags 0 and 1 are used (0 indicates perfect and 1 indicates good) (Fig. 2), and an interpolation method is used to create uniform data in a grid format.

    Fig  2.  FY-4B/GIIRS temperature at 500 hPa at 0100 UTC 23 December 2023.

    The Integrated Global Radiosonde Archive (IGRA) version 2 consists of quality-controlled radiosonde observations of temperature, humidity, and wind at stations across all continents. Data are drawn from more than 30 different sources including over 2800 sounding stations worldwide. The earliest year of data is 1905, and the data are updated on a daily basis (Durre et al., 2018).

    In this study, meteorological radiosonde station observation data are used to validate the accuracy of satellite-derived AMVs and temperature. The locations, station numbers (01–56), and terrain heights of the selected radiosonde stations are shown in Fig. 3. These include 25 international and 31 Chinese meteorological radiosonde stations, totaling 56 stations. Detailed information about the stations can be found in Table 1. The data cover the period from January to December 2023 at 0000 UTC and 1200 UTC.

    Fig  3.  Elevation (shaded), NECV activity region (blue box; 35°–60°N, 105°–145°E), and locations of meteorological radiosonde stations (blue for international and red for Chinese stations).
    Table  1.  Meteorological radiosonde station information. The number (01–56) and location (latitude: °N, longitude: °E) are shown in Fig. 3
    Country Number Station Location Country Number Station Location
    Russia 01 30230 (57.77, 108.07) China 29 54727 (36.65, 117.52)
    Russia 02 31004 (58.60, 125.39) China 30 54857 (36.07, 120.33)
    Russia 03 30635 (53.42, 109.02) China 31 54778 (37.15, 122.37)
    Russia 04 30935 (50.37, 108.76) China 32 53513 (40.72, 107.37)
    Russia 05 30557 (54.43, 113.59) China 33 53336 (41.57, 108.51)
    Russia 06 30758 (52.08, 113.48) China 34 53543 (39.82, 110.01)
    Russia 07 30673 (53.72, 119.76) China 35 53463 (40.86, 111.57)
    Russia 08 31510 (50.28, 127.48) China 36 53038 (43.63, 111.94)
    Russia 09 32061 (50.90, 142.17) China 37 54102 (43.95, 116.12)
    Russia 10 32098 (49.22, 143.10) China 38 54218 (42.31, 118.83)
    Russia 11 31770 (49.00, 140.30) China 39 54135 (43.60, 122.26)
    Russia 12 32150 (46.95, 142.72) China 40 50834 (46.60, 121.21)
    Russia 13 31873 (45.88, 133.72) China 41 50527 (49.25, 119.70)
    Russia 14 31977 (43.26, 132.05) China 42 50557 (49.16, 125.23)
    Mongolia 15 44292 (47.92, 106.87) China 43 50745 (47.38, 123.92)
    Democratic People’s Republic of Korea 16 47102 (37.97, 124.71) China 44 50953 (45.93, 126.57)
    Republic of Korea 17 47104 (37.80, 128.85) China 45 50774 (47.71, 128.84)
    Republic of Korea 18 47230 (37.26, 126.10) China 46 54161 (43.89, 125.23)
    Republic of Korea 19 47138 (36.03, 129.38) China 47 54292 (42.87, 129.50)
    Republic of Korea 20 47155 (35.17, 128.57) China 48 54374 (41.80, 126.89)
    Japan 21 47401 (45.42, 141.68) China 49 54497 (40.03, 124.33)
    Japan 22 47412 (43.06, 141.33) China 50 54218 (41.13, 121.12)
    Japan 23 47582 (39.72, 140.10) China 51 54662 (38.91, 121.64)
    Japan 24 47600 (37.39, 136.90) China 52 54401 (40.77, 114.92)
    Japan 25 47646 (36.06, 140.13) China 53 54539 (39.43, 118.89)
    China 26 53915 (35.53, 106.66) China 54 53798 (37.18, 114.36)
    China 27 53614 (38.47, 106.21) China 55 54511 (39.81, 116.47)
    China 28 53845 (36.58, 109.45) China 56 53772 (37.62, 112.58)
     | Show Table
    DownLoad: CSV

    The reanalysis geopotential height, wind field, and temperature data used in this study are from the ECMWF reanalysis data version 5 (ERA5) (Hersbach et al., 2020). This dataset integrates model and global observation data, with a horizontal spatial resolution of 0.25° (longitude) × 0.25° (latitude) and a temporal resolution of 1 h. The data were downloaded from https://cds.climate.copernicus.eu/.

    The calculations for mean bias (MB), mean absolute error (MAE), root mean square error (RMSE), and Pearson correlation coefficient (CC) are given by Eqs. (1)–(4) below (Zhu, 2023):

    MB=1nni=1(yixi), (1)
    MAE=1nni=1|yixi|, (2)
    RMSE=1nni=1(yixi)2, (3)
    CC=Cov(x,y)sxsy=1n1ni=1[(xi¯x)(yi¯y)][1n1ni=1(xi¯x)2]1/2[1n1ni=1(yi¯y)2]1/2=ni=1[(xi¯x)(yi¯y)][ni=1(xi¯x)2]1/2[ni=1(yi¯y)2]1/2, (4)

    where y is the variable being tested, x is the reference variable, n is the number of matching samples, ¯y is the mean of the tested variable for n samples, and ¯x is the mean of the reference variable for n samples.

    The regression line is characterized by the equations presented in Eqs. (5)–(7) below:

    ˆy=a+bx, (5)
    b=ni=1[(xi¯x)(yi¯y)]ni=1(xi¯x)2, (6)
    a=¯yb¯x, (7)

    where a is the intercept, b is the slope, and the circumflex (“hat”) accent signifies that the equation specifies a predicted value of y.

    (1) Spatial matching method

    Within the coverage area of the FY-4B satellite product, the nearest meteorological radiosonde station is searched for horizontal spatial matching, with a maximum spatial matching distance of 20 km. In the vertical direction, the data-matching pressure range is ±5 hPa.

    (2) Temporal matching method

    Using the radiosonde observations at 0000 and 1200 UTC as the reference, the FY-4B AMV data are matched at 0000 and 1200 UTC. The average of the FY-4B/GIIRS temperature at 2300 UTC of the previous day and 0100 UTC of the current day is matched with the radiosonde temperature at 0000 UTC, and the average of the temperature at 1100 and 1300 UTC is matched with the radiosonde temperature at 1200 UTC.

    Four channels’ AMVs from FY-4B are used in tracking the NECV vortex center. The priority order of the channels is determined through an accuracy evaluation of the AMVs. Based on the statistical analysis of the AMVs’ vertical distribution, the pressure levels of the AMVs used to track the NECV center are determined. The uniform wind field in grid format is created. When tracking the NECV vortex center using AMVs, the “8-point” method is adopted (Zhang, 2001; Huang and Li, 2020). Requirements have been set for the wind direction in the east, south, west, and north sides of the low-value center to better ensure that it meets the characteristics of cyclonic circulation, as shown in Fig. 4. The temperature at 500 hPa from FY-4B/GIIRS is also introduced in tracking the NECV center.

    Fig  4.  Diagram for identifying the cyclonic circulation of the NECV: u and v are the latitudinal and meridional components of the AMV, respectively (Huang and Li, 2020).

    (1) FY-4B/AGRI AMV

    Using wind data from the IGRA, the accuracy of FY-4B AMVs’ wind speed and wind direction data from four channels was evaluated separately. From January to December 2023, the spatiotemporal matching samples for CH09, CH10, CH11, and CH13 are 7739, 9268, 8860, and 6060, respectively. The scatter density plots and evaluation metrics for wind speed (Fig. 5) show that the high-density scatter regions of the four channels are located in different wind speed ranges. The samples obtained from CH09 are concentrated in the higher wind speed range of 10–25 m s−1, while those from CH13 are concentrated in the wind speed range below 15 m s−1. The different dense scatter regions represent the vertical height distribution of the AMVs retrieved from the four channels. Generally, in the mid–high-latitude westerlies, the average wind speed increases with height from the lower to the upper troposphere. Comparing the wind speed evaluation indices of the four channels shows that CH10 has the highest CC (0.89) and the smallest MAE (4.7 m s−1) and RMSE (7.07 m s−1). CH09’s CC is slightly lower than that of CH10 at 0.87, and CH11 has the lowest CC at 0.67.

    Fig  5.  Scatter density and evaluation indices of wind speed (m s−1) between FY-4B AMV and IGRA wind speed data for the NECV activity area ( 35°–60°N, 105°–145°E) for (a) CH09, (b) CH10, (c) CH11, and (d) CH13 (red solid line is regression line).

    The scatter density plots and evaluation metrics for wind direction (Fig. 6) show that the high-density scatter regions for wind direction in the NECV activity area are concentrated between 250° and 300°. Among the four channels, the minimum CC for wind direction is 0.92 (CH10 and CH11), with CH09 having a CC of 0.94. CH09 and CH10 have a smaller wind direction MAE at 10.56° and 11.25°, respectively, while CH11 has the largest at 12.09°. Among the four channels, CH09 has the smallest wind direction RMSE at 17.42°, and CH11 has the largest at 21.44°. The wind direction RMSE for CH09 is the smallest, but for wind speed, all evaluation indicators for CH10 are better than those of CH09 (Fig. 5). In the wind direction evaluation, the MB of CH10 is also better than that of CH09. Overall, including the sample number of wind vectors, the priority order selected is CH10 and CH09. In summary, based on the evaluation results of wind speed and wind direction for the four channels, when identifying the NECV, the priority order for using the AMV data from the four channels at the same time is CH10, CH09, CH13, and CH11.

    Fig  6.  As in Fig. 5, but for wind direction.

    The vertical distribution shows good AMV data quality from 200 to 400 hPa with a relative high CC and small RMSE. The maximum wind speed/direction CC is 0.89/0.80 at 200 hPa (Table 2).

    Table  2.  FY-4B AMV accuracy at different vertical levels
    Level (hPa) CC MB (m s−1) MAE (m s−1) RMSE (m s−1) CC MB (°) MAE (°) RMSE (°)
    Wind speed Wind direction
    100 0.58 −1.68 4.00 5.27 0.56 −6.24 25.23 45.74
    200 0.89 −3.34 4.69 6.23 0.80 −4.12 11.46 24.10
    300 0.85 −1.52 4.67 7.17 0.77 −8.71 18.71 47.28
    400 0.67 −1.35 4.88 7.42 0.77 −2.58 25.61 46.77
    500 0.37 −2.18 5.36 9.51 0.76 −9.73 29.82 59.82
     | Show Table
    DownLoad: CSV

    (2) FY-4B/GIIRS temperature

    Using IGRA temperature data, an assessment of the accuracy of the FY-4B/GIIRS temperature was conducted. The selected vertical layers of the FY-4B/GIIRS temperature were from the 45th (103 hPa) to the 95th layer (932 hPa). From January to December 2023, the total number of matching samples is 127,307. The scatter density plots and evaluation metrics (Fig. 7) show that the CC (0.996) between the two datasets is relatively high, the MB is −0.26 K, indicating that the FY-4B/GIIRS temperature is slightly lower than the radiosonde data. The MAE is 1.43 K and the RMSE is 1.87 K. The accuracy evaluation results are better than those of FY-4A/GIIRS (Ren et al., 2022). The vertical distribution of the evaluation results shows that the data quality at 500 hPa is relatively high, with minimum MAE (1.07 K) and RMSE (1.41 K) (Table 3). The temperature profile can be used not only in tracking the cold center of the NECV but also in analyzing the three-dimensional thermal structure of the atmosphere.

    Fig  7.  Scatter density and evaluation indices of FY-4B/GIIRS and IGRA temperature in the NECV activity area (35°–60°N, 105°–145°E) (red solid line: regression line).
    Table  3.  FY-4B/GIIRS temperature accuracy at different vertical levels
    Level (hPa)RRMB (K)MAE (K)RMSE (K)
    1000.92−0.501.612.15
    2000.93−0.271.561.97
    3000.97−0.381.251.71
    4000.99−0.461.131.52
    5000.99−0.261.071.41
    6000.990.021.211.54
    7000.990.071.251.55
    8500.981.152.012.60
    9250.971.242.102.79
     | Show Table
    DownLoad: CSV

    Using meteorological radiosonde data and reanalysis data, the tracking of the NECV center generally involves the mid–upper troposphere geopotential height, wind, or temperature (Nieto et al., 2005). The AMV retrieved from the water vapor channel of the satellite mostly represents the mid–upper troposphere wind, while the AMV retrieved from the infrared channel involves some of the mid–lower troposphere wind based on the cloud top height. To determine the pressure layer range of the AMV used for tracking the NECV center, a vertical distribution statistical analysis of the wind sample number retrieved from the four channels was conducted (Fig. 8). The weight altitude is the highest for CH09, with the largest percentage of AMV samples around 260 hPa at approximately 13.2%, followed by the neighboring 240 and 280 hPa, where the AMV samples are mostly concentrated. AMV samples exceeding 1.0% are concentrated from 180 to 400 hPa in the mid–upper troposphere, accounting for 93.5% in total. Most of the AMV samples are found at 300 and 320 hPa of the water vapor channels CH10 and CH11, respectively. The AMV samples of CH10 exceeding 1.0% are from 160 to 520 hPa, with a larger range in the vertical direction, accounting for 93.4% in total. The AMV samples of CH11 exceeding 1.0% are from 200 to 600 hPa, with an even larger vertical range, accounting for 91.5% in total. The AMV samples of the infrared channel CH13 have the largest number in the mid troposphere at around 500 hPa, with AMVs exceeding 1.0% from 200 to 820 hPa, accounting for 96.3% in total, making this channel with the highest percentage of mid–lower-troposphere AMVs among the four channels. The average analysis of the four channels (black line) shows that AMVs exceeding 2.0% (1.0%) are from 200 to 500 hPa (200 to 640 hPa), accounting for 83.1% (89.0%) in total. When identifying the center of the NECV, it is necessary to obtain as many AMV samples as possible within a small pressure layer range in the mid–upper troposphere. Based on the vertical distribution characteristics of the AMV, the winds from 200 to 500 hPa of the four channels are selected for identifying the NECV center.

    Fig  8.  Vertical distributions of FY-4B AMV number percentage of CH09, CH10, CH11, CH13, and average of the four channels.

    The NECV can occur throughout the year. In summer, it is often accompanied by severe convective weather, frequently bringing thunderstorms, strong winds, heavy rainfall, and hail to regions such as Nei Mongol Zizhiqu, Northeast China, North China, and the Huang–Huai and Jiang–Huai regions (Zhang et al., 2019). In winter, the NECV is related to extreme low-temperature events in China (Yin et al., 2013), often causing rain, snow, and severe freezing. In this study, NECV events in 2023 were chosen for an overall evaluation of the differences in the NECV center data from different datasets. Two strong NECV events were also selected respectively in summer and winter 2023, to analyze the application of FY-4B AMV and temperature products in tracking the NECV center in different seasons. The time periods of the two NECV events are from 3 to 10 July and from 4 to 7 November 2023.

    (1) Overview of the two NECV events

    An NECV event occurred from 3 to 10 July 2023. According to the true color images from the FY-3D meteorological satellite on 4 July (Fig. 9a), the center of the NECV was located in eastern Mongolia, with spiral cloud bands on the southeastern side affecting eastern Nei Mongol Zizhiqu, Northeast China, eastern North China, and the eastern Huang–Huai region. On 7 July (Fig. 9b), the center of the NECV had moved into China, with large areas of scattered convective cloud clusters developing on the eastern and southern sides of the vortex cloud system. In November 2023, China experienced multiple cold-wave weather events accompanied by strong winds, temperature drops, and blizzards. From 4 to 7 November, affected by the NECV, extreme rain and snow occurred in northern North China, Nei Mongol Zizhiqu, and Northeast China. Daily precipitation records at 41 national meteorological stations in Heilongjiang and Liaoning provinces broke historical records in November. According to the true color images from the FY-3D meteorological satellite, on 5 November (Fig. 9c), the center of the NECV was located in the northern part of western Nei Mongol Zizhiqu, with the vortex cloud system on the eastern and southeastern sides affecting central and eastern Nei Mongol Zizhiqu, Northeast China, North China, Huang–Huai, Jiang–Huai, and Jiang–Han regions. On 6 November (Fig. 9d), as the cold vortex gradually moved eastward, the cloud system mainly covered central and eastern Nei Mongol Zizhiqu and Northeast China, causing heavy snowfall.

    Fig  9.  True color images of the NECV from the FY-3D satellite on (a) 4 July, (b) 7 July, (c) 5 November, and (d) 6 November 2023.

    (2) Identify the NECV center using FY-4B AMV

    The 200–500-hPa range represents the mid–upper troposphere. Using the wind field from FY-4B, the center of the NECV was identified and compared with the low-pressure center from the ERA5 geopotential height at 500 hPa (Fig. 10, Fig. 11, and Table 4). The tracks of the NECV identified from both datasets are almost consistent with the AMV vortex center slightly to the west–southwest of the low-pressure center in the ERA5 geopotential height field.

    Fig  10.  (a–f) FY-4B AMV (barbs, wind field in the range of 200–500 hPa) and ERA5 geopotential height at 500 hPa (shaded, white contour intervals of 2, unit: 10 gpm) at 0000 UTC from 4 to 9 July 2023 during the NECV event. The red squares and lines indicate the position and path of the AMV vortex center, while the blue dots and lines indicate the position and path of the low-pressure center of the geopotential height at 500 hPa.
    Fig  11.  As in Fig. 10, but for the NECV event from 4 to 7 November 2023. (a) 1200 UTC 4 November; (b) 1200 UTC 5 November; (c) 0000 UTC 6 November; (d) 1200 UTC 6 November; (e) 0000 UTC 7 November; and (f) 1200 UTC 7 November.
    Table  4.  NECV center positions (latitude: °N, longitude: °E) from different datasets [(a) FY-4B AMV at 200–500 hPa, (b) ERA5 geopotential height at 500 hPa, (c) radiosonde observation, (d) FY-4B/GIIRS temperature at 500 hPa, and (e) ERA5 temperature at 500 hPa] and their difference (distance; km)
    Time (mm-dd UTC)(a)(b)(c)Distance (a)(b)Distance (b)(c)(d)(e)Distance (d)(e)
    07-03 1200(45.9, 103.5)(46.9, 105.6)195.7(45.5, 104.0)(45.0, 104.7)78.1
    07-04 0000(46.3, 108.0)(48.1, 109.5)(48, 107)230.0186.2(48.0, 108.1)(48.2, 108.8)56.5
    07-04 1200(46.8, 110.2)(47.9, 111.3)(49, 110)147.7155.4(48.9, 110.2)(48.7, 110.3)23.4
    07-05 0000(46.0, 114.2)(47.0, 114.5)(48, 115)113.5117.4(48.1, 113.0)(47.2, 112.5)106.9
    07-05 1200(44.6, 121.0)(46.8, 123.0)(47, 123)289.722.2(46.0, 116.0)(45.4, 116.0)66.7
    07-06 0000(44.4, 121.8)(47.2, 119.0)(47, 120)379.578.9(45.0, 125.5)(45.5, 122.7)226.1
    07-06 1200(45.0, 115.5)(46.7, 115.6)(47, 122)189.2487.7(45.4, 127.2)(45.5, 133.0)452.5
    07-07 0000(44.0, 114.8)(45.0, 115.2)(44, 116)115.6128.0(44.0, 129.5)(42.5, 131.5)232.5
    07-07 1200(44.0, 115.5)(44.5, 116.2)(45, 116)78.757.8(43.5, 131.0)(42.4, 130.6)126.6
    07-08 0000(43.5, 118.3)(45.0, 119.5)(44, 118)192.2162.8(44.2, 131.2)(43.0, 139.0)641.8
    07-08 1200(45.0, 117.5)(46.4, 118.6)(47, 121)177.6194.8(44.0, 131.4)(44.5, 130.8)73.3
    07-09 0000(42.5, 119.0)(44.3, 121.3)273.1(46.2, 133.2)(46.0, 134.0)65.6
    11-04 1200(44.8, 102.8)(46.0, 104.8)205.4(45.0, 102.5)(45.3, 101.9)57.7
    11-05 0000(44.3, 106.4)(44.7, 107.5)97.9(45.2, 105.2)(45.2, 105.7)39.2
    11-05 1200(43.0, 110.8)(43.5, 111.8)(44, 112)98.257.9(42.5, 110.8)(41.7, 110.9)89.3
    11-06 0000(38.6, 118.2)(39.6, 118.3)(40, 118)111.551.3(40.4, 114.8)(40.1, 116.4)139.8
    11-06 1200(42.3, 126.8)(43.3, 127.2)(41, 127)115.9256.3(43.1, 128.3)(43.3, 127.8)46.2
    11-07 0000(45.2, 132.0)(46.2, 132.5)(46, 133)117.844.5(45.7, 133.6)(46.0, 134.3)63.7
    11-07 1200(47.6, 137.2)(49.5, 138.5)231.9(47.0, 137.5)(48.6, 139.0)210.2
    Average176.9142.9147.2
     | Show Table
    DownLoad: CSV

    During the summer NECV event (Fig. 10), at 1200 UTC 3 July 2023, the 500-hPa low-pressure center entered the NECV activity region (blue rectangle in Fig. 10), with the AMV vortex center slightly to the southwest (the distance is 195.7km), with longitude and latitude deviations of −2.1° and −1.0°, respectively. At 0000 UTC 4 July, the cold vortex center moved northeastward and then turned southeastward. At 0000 UTC 5 July, the 500-hPa geopotential height reached the lowest value of 550 dagpm (1 dagpm = 10 gpm) during this NECV event, with the AMV vortex center and low-pressure center having longitude and latitude deviations of −0.3° and −1.0°, respectively, showing relatively small differences with the distance of 113.5 km. From 6 to 7 July, both the AMVs and the geopotential height field indicated that the NECV center first moved eastward and then shifted southwestward, with relatively large differences in the identified center positions during significant path changes. The largest deviations (the distance was 379.5 km) occurred at 0000 UTC 6 July, with longitude and latitude deviations of 2.8° and −2.4°, respectively. From 1200 UTC 7 July, the NECV center began to rotate and rapidly weakened, dissipating on 9 July. During the rotation, the differences between the AMV vortex center and the low-pressure center were relatively small (the smallest distance was 78.7 km), with larger differences at 0000 UTC 9 July (273.1 km). During the summer NECV event, the average distance between the AMV vortex center and the low-pressure center was 198.5 km. The NECV center differences between radiosondes (provided by Liaoning Meteorological Service) and ERA5 were smaller at 1200 UTC 5 July and at 0000 UTC 6 July, but were bigger at 0000 UTC 7 July.

    During the winter NECV event (Fig. 11), at 1200 UTC 4 November 2023, the low-pressure center at 500 hPa approached the NECV activity region, with the AMV vortex center slightly to the southwest with a distance of 205.4 km, with longitude and latitude deviations of −2.0° and −1.2°, respectively. At 0000 UTC 6 November, the movement direction changed from southeastward to northeastward, reaching the strongest intensity at 1200 UTC 6 November with a minimum geopotential height of 528 dagpm at 500 hPa. Compared to the summer NECV event from 4 to 9 July, the winter NECV event had a faster movement speed and a shorter lifespan. During the lifecycle, the differences between the AMV vortex center and the low-pressure center were relatively large during the generation (at 1200 UTC 4 November) and dissipation stages (at 1200 UTC 7 November), with smaller differences at other times (on 5 November). During this winter NECV event, the average distance between the AMV vortex center and the low-pressure center was 139.8 km which is smaller than that (198.5 km) during the summer NECV event from 4 to 9 July.

    (3) Identify the NECV center using FY-4B/GIIRS temperature

    The NECV is characterized by a cold-core structure throughout the troposphere and a cold center located in the mid–upper troposphere (Liu et al., 2017; Wang et al., 2017). Therefore, the FY-4B/GIIRS temperature at 500 hPa can be used to identify the cold center of the NECV. Figures 12 and 13 show the cold centers identified from the FY-4B/GIIRS and ERA5 temperatures at 500 hPa during the NECV events from 4 to 9 July and from 4 to 7 November 2023, respectively. In summer, due to the influence of the well-developed cloud system of the NECV, the FY-4B/GIIRS temperature data are missing for some regions, especially north of 55°N, where the temperature coverage is sparse, making it difficult to obtain a uniformly distributed temperature field. In winter, the cloud system of the NECV is relatively weak, with a lower cloud top height, allowing for better coverage of data at 500 hPa, this is advantageous for identifying the cold center of the NECV.

    Fig  12.  (a–f) FY-4B/GIIRS (shaded) and ERA5 (white contours) temperature at 500 hPa (°C) during the NECV event at 0000 UTC from 4 to 9 July 2023. The red squares and lines indicate the position and path from the FY-4B/GIIRS cold center, while the green dots and lines indicate the position and path from the ERA5 cold center.
    Fig  13.  As in Fig. 12, but for the NECV event from 4 to 7 November 2023. (a) 1200 UTC 4 November; (b) 1200 UTC 5 November; (c) 0000 UTC 6 November; (d) 1200 UTC 6 November; (e) 0000 UTC 7 November; and (f) 1200 UTC 7 November.

    During the summer NECV event (Fig. 12), before 0000 UTC 6 July 2023, the cold center identified from the temperature and the vortex center identified from the AMVs showed little difference in position. When the vortex center of the NECV began to recirculate, the original cold center continued to move eastward, with cold air replenishing from the west, forming an east–west band of low temperature, resulting in a significant difference between the cold center and the vortex center. From 1200 UTC 3 July to 1200 UTC 5 July, the position difference between the FY-4B and ERA5 cold centers at 500 hPa was small, with distances of 78.1, 56.5, 23.4, 106.9, and 66.7 km, respectively (Table 4). During the recirculation period of the NECV center from 0000 UTC 6 July to 0000 UTC 8 July, the differences between the two were larger. The average distance is 179.2 km.

    During the winter NECV event (Fig. 13), from 4 to 7 November 2023, the position differences between the FY-4B/GIIRS and ERA5 cold centers at 500 hPa were small, with an average distance of 92.3 km. The maximum difference occurred at 1200 UTC 7 November, with a distance of 210.2 km. In this NECV event, the vortex center position identified from AMVs was highly consistent with the cold center position at 500 hPa.

    There were 24 NECV events in 2023, according to the data provided by Liaoning Meteorological Service, and 192 center positions were examined (center positions at 0000 and 1200 UTC). The results are shown in Fig. 14. The average distance is 181.9 km between the center positions identified from the upper troposphere FY-4B AMVs and ERA5 geopotential height at 500 hPa; the average distance is 140.6 km between the center positions identified from the FY-4B and ERA5 temperatures at 500 hPa, showing similar results to the two typical NECV events in summer and winter in 2023.

    Fig  14.  NECV center position difference (distance: km) from different datasets for 192 cases in 2023 (blue line, distance between upper troposphere FY-4B AMV and ERA5 geopotential height at 500 hPa; red line, distance between FY-4B and ERA5 temperature at 500 hPa).

    The findings obtained thus far indicate the potential for a seasonal strategy—specifically, utilizing the reduced cloud cover during winter to facilitate broader temperature retrievals and capitalizing on summer’s dominance of AMV—to improve the reliability of NECV tracking. This nuanced approach has not been addressed in previous studies.

    This study examined the application of the FY-4B satellite-retrieved AMVs and temperature products in monitoring the NECV. Firstly, a data accuracy evaluation was conducted in the NECV activity region. The vertical distribution characteristics of the AMVs were studied, and methods for identifying the center of the NECV using AMVs and temperature were discussed. The applicability of satellite products in identifying the NECV was analyzed for two NECV cases: one in summer and the other in winter. The main conclusions are as follows.

    (1) In the NECV activity region, compared with meteorological sounding observations, the wind speed evaluation showed that CH10 had the highest CC (0.87) and the lowest MAE (4.7 m s−1) and RMSE (7.07 m s−1). CH09 had a slightly lower CC than CH10, and CH11 had the lowest (0.67). The wind direction scatter density in the NECV activity region was concentrated between 250° and 300°. The wind direction MAEs for CH09 and CH10 were 10.56° and 11.25°, and the RMSEs were 17.42° and 20.57°, respectively. In identifying the NECV, the priority order for using AMV channel data is CH10, CH09, CH13, and CH11.

    (2) The temperature evaluation of the FY-4B/GIIRS in the NECV activity region showed that, compared with meteorological sounding observations, the CC of the two datasets was relatively high (0.996). The MB was −0.26 K, indicating that the FY-4B/GIIRS temperature was slightly lower. The MAE was 1.43 K and the RMSE was 1.87 K.

    (3) When identifying the center of the NECV, it is necessary to obtain as many AMV quantities as possible, which are concentrated in a small pressure layer range in the mid–upper troposphere. Based on the vertical distribution characteristics of the AMVs, four channels of AMV data from 200 to 500 hPa were selected for identifying the center of the NECV, accounting for 83.1% of the total data samples.

    (4) When AMVs were used to identify the vortex center in the mid–upper troposphere of the NECV, the tracks from the AMV vortex center were generally consistent with the low-pressure center from the ERA5 geopotential height at 500 hPa, with the center position being slightly biased to the west and south. During the summer/winter NECV events, the average distance between the AMV vortex center and the low-pressure center was 198.5/139.8 km, with relatively smaller deviations in winter. The average distance was 181.9 km for the 24 NECV events in 2023.

    (5) When the FY-4B/GIIRS temperature at 500 hPa was used to identify the cold center of the NECV, compared with the ERA5 temperature at 500 hPa, the average distance for the summer/winter case was 179.2/92.3 km. The average distance was 140.6 km for the 24 NECV events in 2023. Due to the characteristics of FY-4B/GIIRS’s infrared detection, there was higher spatial coverage in winter when cloud development was weaker, and the absolute deviation in identifying the cold center was smaller compared to the reanalysis data.

    In general, this study (i) elucidates the vertical distribution of AMV utility within the 200–500-hPa range and addresses seasonal retrieval challenges, thereby enhancing our understanding the limitations of satellite data in vortex studies; (ii) establishes the accuracy of temperature retrieval from FY-4B/GIIRS (CC = 0.996, RMSE = 1.87 K) as a benchmark for future applications of geostationary sounders in midlatitude systems; (iii) validates satellite-derived NECV centers against ERA5 reanalysis and radiosonde data, quantifying positional errors (e.g., mean distance: 140–180 km), an aspect that previous NECV studies have seldom addressed; and (IV) provides a real-time monitoring framework for NECV using high-frequency data from FY-4B, which is critical for early warnings related to severe weather events such as convective storms and extreme cold linked to NECV activities.

    This study mainly focused on the application potential of FY-4B AMVs and temperature in the tracking of the NECV center. The subjective and objective identification methods of the NECV in other studies are mainly based on meteorological sounding observations and reanalysis data. Meteorological sounding observations have high timeliness and high quality, but the spatial resolution and time frequency are low (twice a day), making it difficult to obtain accurate data on the central position and continuous evolution of the NECV. Reanalysis data exhibit significant time latency and can be utilized for the development and research of historical NECV datasets. However, they are not suitable for real-time operational services. This study utilized quantitative products from FY-4B, which can provide wind field data with 15-min intervals and three-dimensional temperature profiles with 2-h intervals. The aim is to achieve high-frequency observations of the NECV center and provide support for operational disaster weather services related to the NECV. NECV centers are primarily identified from AMV fields, and the cold temperature constraint is only necessary for distinguishing them from thermal low vortices. In the baroclinic developing stage, the cold center and low vortex center are never in the same place.

    There are seasonal differences in the application of satellite products. In warm seasons, there is greater cloud development during NECV activity, and the changes are more obvious in infrared cloud images and water vapor images from which more AMVs are derived. Therefore, AMVs have more advantages in terms of tracking the center of the NECV in summer. In cold seasons, due to relatively weak cloud activity and uniform water vapor grayscale, the number of retrieved AMVs is relatively low. There are generally more AMVs near the center of the cold vortex than other regions, providing information for tracking the vortex center in winter (Fig. 11). Further research is required on objective identification methods based on satellite-derived AMV and temperature data, and an analysis of the accuracy of long-term series identification also needs to be carried out to achieve the goal of operational applications.

    On 5 March 2024, the FY-4B satellite drifted from 133°E to 105°E, replacing the FY-4A satellite. Figure 15 shows the spatial coverage of the temperature data before and after the adjustment of the FY-4B sub-satellite point. After the adjustment, the coverage of the FY-4B/GIIRS temperature further extended westward on the western side and shifted slightly westward in the high-latitude areas on the eastern side. This can provide a better upstream observation advantage for the NECV activity region. The FY-4B/GIIRS includes products that measure not only temperature, but also humidity. The combined application of temperature and humidity measurement devices can facilitate the study and real-time monitoring of the structure of the NECV, as well as the atmospheric stratification and energy characteristics of hazardous weather triggered by the cold vortex. In addition to tracking the NECV, AMVs can also provide information on the dynamic characteristics of the upper troposphere (such as divergence and upper-level jets), which can be used for the monitoring of precipitation intensity and distribution, severe thunderstorms, and other hazardous weather associated with the NECV in a timely manner.

    Fig  15.  FY-4B/GIIRS temperature at 500 hPa at (a) 1100 UTC 4 November 2023 (sub-satellite point at 133°E) and (b) 0100 UTC 7 March 2024 (sub-satellite point at 105°E).
  • Fig.  10.   (a–f) FY-4B AMV (barbs, wind field in the range of 200–500 hPa) and ERA5 geopotential height at 500 hPa (shaded, white contour intervals of 2, unit: 10 gpm) at 0000 UTC from 4 to 9 July 2023 during the NECV event. The red squares and lines indicate the position and path of the AMV vortex center, while the blue dots and lines indicate the position and path of the low-pressure center of the geopotential height at 500 hPa.

    Fig.  1.   FY-4B AMVs from (a) CH09, (b) CH10, (c) CH11, and (d) CH13 at 0600 UTC 7 July 2023.

    Fig.  2.   FY-4B/GIIRS temperature at 500 hPa at 0100 UTC 23 December 2023.

    Fig.  3.   Elevation (shaded), NECV activity region (blue box; 35°–60°N, 105°–145°E), and locations of meteorological radiosonde stations (blue for international and red for Chinese stations).

    Fig.  4.   Diagram for identifying the cyclonic circulation of the NECV: u and v are the latitudinal and meridional components of the AMV, respectively (Huang and Li, 2020).

    Fig.  5.   Scatter density and evaluation indices of wind speed (m s−1) between FY-4B AMV and IGRA wind speed data for the NECV activity area ( 35°–60°N, 105°–145°E) for (a) CH09, (b) CH10, (c) CH11, and (d) CH13 (red solid line is regression line).

    Fig.  6.   As in Fig. 5, but for wind direction.

    Fig.  7.   Scatter density and evaluation indices of FY-4B/GIIRS and IGRA temperature in the NECV activity area (35°–60°N, 105°–145°E) (red solid line: regression line).

    Fig.  8.   Vertical distributions of FY-4B AMV number percentage of CH09, CH10, CH11, CH13, and average of the four channels.

    Fig.  9.   True color images of the NECV from the FY-3D satellite on (a) 4 July, (b) 7 July, (c) 5 November, and (d) 6 November 2023.

    Fig.  11.   As in Fig. 10, but for the NECV event from 4 to 7 November 2023. (a) 1200 UTC 4 November; (b) 1200 UTC 5 November; (c) 0000 UTC 6 November; (d) 1200 UTC 6 November; (e) 0000 UTC 7 November; and (f) 1200 UTC 7 November.

    Fig.  12.   (a–f) FY-4B/GIIRS (shaded) and ERA5 (white contours) temperature at 500 hPa (°C) during the NECV event at 0000 UTC from 4 to 9 July 2023. The red squares and lines indicate the position and path from the FY-4B/GIIRS cold center, while the green dots and lines indicate the position and path from the ERA5 cold center.

    Fig.  13.   As in Fig. 12, but for the NECV event from 4 to 7 November 2023. (a) 1200 UTC 4 November; (b) 1200 UTC 5 November; (c) 0000 UTC 6 November; (d) 1200 UTC 6 November; (e) 0000 UTC 7 November; and (f) 1200 UTC 7 November.

    Fig.  14.   NECV center position difference (distance: km) from different datasets for 192 cases in 2023 (blue line, distance between upper troposphere FY-4B AMV and ERA5 geopotential height at 500 hPa; red line, distance between FY-4B and ERA5 temperature at 500 hPa).

    Fig.  15.   FY-4B/GIIRS temperature at 500 hPa at (a) 1100 UTC 4 November 2023 (sub-satellite point at 133°E) and (b) 0100 UTC 7 March 2024 (sub-satellite point at 105°E).

    Table  1   Meteorological radiosonde station information. The number (01–56) and location (latitude: °N, longitude: °E) are shown in Fig. 3

    Country Number Station Location Country Number Station Location
    Russia 01 30230 (57.77, 108.07) China 29 54727 (36.65, 117.52)
    Russia 02 31004 (58.60, 125.39) China 30 54857 (36.07, 120.33)
    Russia 03 30635 (53.42, 109.02) China 31 54778 (37.15, 122.37)
    Russia 04 30935 (50.37, 108.76) China 32 53513 (40.72, 107.37)
    Russia 05 30557 (54.43, 113.59) China 33 53336 (41.57, 108.51)
    Russia 06 30758 (52.08, 113.48) China 34 53543 (39.82, 110.01)
    Russia 07 30673 (53.72, 119.76) China 35 53463 (40.86, 111.57)
    Russia 08 31510 (50.28, 127.48) China 36 53038 (43.63, 111.94)
    Russia 09 32061 (50.90, 142.17) China 37 54102 (43.95, 116.12)
    Russia 10 32098 (49.22, 143.10) China 38 54218 (42.31, 118.83)
    Russia 11 31770 (49.00, 140.30) China 39 54135 (43.60, 122.26)
    Russia 12 32150 (46.95, 142.72) China 40 50834 (46.60, 121.21)
    Russia 13 31873 (45.88, 133.72) China 41 50527 (49.25, 119.70)
    Russia 14 31977 (43.26, 132.05) China 42 50557 (49.16, 125.23)
    Mongolia 15 44292 (47.92, 106.87) China 43 50745 (47.38, 123.92)
    Democratic People’s Republic of Korea 16 47102 (37.97, 124.71) China 44 50953 (45.93, 126.57)
    Republic of Korea 17 47104 (37.80, 128.85) China 45 50774 (47.71, 128.84)
    Republic of Korea 18 47230 (37.26, 126.10) China 46 54161 (43.89, 125.23)
    Republic of Korea 19 47138 (36.03, 129.38) China 47 54292 (42.87, 129.50)
    Republic of Korea 20 47155 (35.17, 128.57) China 48 54374 (41.80, 126.89)
    Japan 21 47401 (45.42, 141.68) China 49 54497 (40.03, 124.33)
    Japan 22 47412 (43.06, 141.33) China 50 54218 (41.13, 121.12)
    Japan 23 47582 (39.72, 140.10) China 51 54662 (38.91, 121.64)
    Japan 24 47600 (37.39, 136.90) China 52 54401 (40.77, 114.92)
    Japan 25 47646 (36.06, 140.13) China 53 54539 (39.43, 118.89)
    China 26 53915 (35.53, 106.66) China 54 53798 (37.18, 114.36)
    China 27 53614 (38.47, 106.21) China 55 54511 (39.81, 116.47)
    China 28 53845 (36.58, 109.45) China 56 53772 (37.62, 112.58)
    Download: Download as CSV

    Table  2   FY-4B AMV accuracy at different vertical levels

    Level (hPa) CC MB (m s−1) MAE (m s−1) RMSE (m s−1) CC MB (°) MAE (°) RMSE (°)
    Wind speed Wind direction
    100 0.58 −1.68 4.00 5.27 0.56 −6.24 25.23 45.74
    200 0.89 −3.34 4.69 6.23 0.80 −4.12 11.46 24.10
    300 0.85 −1.52 4.67 7.17 0.77 −8.71 18.71 47.28
    400 0.67 −1.35 4.88 7.42 0.77 −2.58 25.61 46.77
    500 0.37 −2.18 5.36 9.51 0.76 −9.73 29.82 59.82
    Download: Download as CSV

    Table  3   FY-4B/GIIRS temperature accuracy at different vertical levels

    Level (hPa)RRMB (K)MAE (K)RMSE (K)
    1000.92−0.501.612.15
    2000.93−0.271.561.97
    3000.97−0.381.251.71
    4000.99−0.461.131.52
    5000.99−0.261.071.41
    6000.990.021.211.54
    7000.990.071.251.55
    8500.981.152.012.60
    9250.971.242.102.79
    Download: Download as CSV

    Table  4   NECV center positions (latitude: °N, longitude: °E) from different datasets [(a) FY-4B AMV at 200–500 hPa, (b) ERA5 geopotential height at 500 hPa, (c) radiosonde observation, (d) FY-4B/GIIRS temperature at 500 hPa, and (e) ERA5 temperature at 500 hPa] and their difference (distance; km)

    Time (mm-dd UTC)(a)(b)(c)Distance (a)(b)Distance (b)(c)(d)(e)Distance (d)(e)
    07-03 1200(45.9, 103.5)(46.9, 105.6)195.7(45.5, 104.0)(45.0, 104.7)78.1
    07-04 0000(46.3, 108.0)(48.1, 109.5)(48, 107)230.0186.2(48.0, 108.1)(48.2, 108.8)56.5
    07-04 1200(46.8, 110.2)(47.9, 111.3)(49, 110)147.7155.4(48.9, 110.2)(48.7, 110.3)23.4
    07-05 0000(46.0, 114.2)(47.0, 114.5)(48, 115)113.5117.4(48.1, 113.0)(47.2, 112.5)106.9
    07-05 1200(44.6, 121.0)(46.8, 123.0)(47, 123)289.722.2(46.0, 116.0)(45.4, 116.0)66.7
    07-06 0000(44.4, 121.8)(47.2, 119.0)(47, 120)379.578.9(45.0, 125.5)(45.5, 122.7)226.1
    07-06 1200(45.0, 115.5)(46.7, 115.6)(47, 122)189.2487.7(45.4, 127.2)(45.5, 133.0)452.5
    07-07 0000(44.0, 114.8)(45.0, 115.2)(44, 116)115.6128.0(44.0, 129.5)(42.5, 131.5)232.5
    07-07 1200(44.0, 115.5)(44.5, 116.2)(45, 116)78.757.8(43.5, 131.0)(42.4, 130.6)126.6
    07-08 0000(43.5, 118.3)(45.0, 119.5)(44, 118)192.2162.8(44.2, 131.2)(43.0, 139.0)641.8
    07-08 1200(45.0, 117.5)(46.4, 118.6)(47, 121)177.6194.8(44.0, 131.4)(44.5, 130.8)73.3
    07-09 0000(42.5, 119.0)(44.3, 121.3)273.1(46.2, 133.2)(46.0, 134.0)65.6
    11-04 1200(44.8, 102.8)(46.0, 104.8)205.4(45.0, 102.5)(45.3, 101.9)57.7
    11-05 0000(44.3, 106.4)(44.7, 107.5)97.9(45.2, 105.2)(45.2, 105.7)39.2
    11-05 1200(43.0, 110.8)(43.5, 111.8)(44, 112)98.257.9(42.5, 110.8)(41.7, 110.9)89.3
    11-06 0000(38.6, 118.2)(39.6, 118.3)(40, 118)111.551.3(40.4, 114.8)(40.1, 116.4)139.8
    11-06 1200(42.3, 126.8)(43.3, 127.2)(41, 127)115.9256.3(43.1, 128.3)(43.3, 127.8)46.2
    11-07 0000(45.2, 132.0)(46.2, 132.5)(46, 133)117.844.5(45.7, 133.6)(46.0, 134.3)63.7
    11-07 1200(47.6, 137.2)(49.5, 138.5)231.9(47.0, 137.5)(48.6, 139.0)210.2
    Average176.9142.9147.2
    Download: Download as CSV
  • Durre, I., X. G. Yin, R. S. Vose, et al., 2018: Enhancing the data coverage in the integrated global radiosonde archive. J. Atmos. Oceanic Technol., 35, 1753–1770, https://doi.org/10.1175/JTECH-D-17-0223.1.
    He, H., Y. Chen, T. G. Xiao, et al., 2015: Statistical analysis of severe short-time precipitation under cold vortex background. Meteor. Mon., 41, 1466–1476. (in Chinese)
    He, J. H., Z. W. Wu, Z. H. Jiang, et al., 2006: The “climate effect” of the Northeast Cold Vortex and its impact on Meiyu rain. Chinese Sci. Bull., 51 , 2803–2809, doi: 10.1360/csb2006-51-23-2803. (in Chinese)
    Hersbach, H., B. Bell, P. Berrisford, et al., 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 1999–2049, https://doi.org/10.1002/qj.3803.
    Huang, X., and D. L. Li, 2020: Objective identification method and variation characteristics of the Northeast China Cold Vortex from May to August of 1979–2018. Acta Meteor. Sinica, 78, 945–961. (in Chinese)
    Jiang, D. K., J. S. Wang, Q. Yan, et al., 2012: Climatic characters of Northeast Cold Vortex and its effect on air temperature in Liaoning province from May to September during 1961–2010. J. Meteor. Environ., 28, 5–9. (in Chinese)
    Lin, Z. Q., W. D. Guo, X. P. Yao, et al., 2023: Reexamine the Tibetan Plateau vortices sources based on multiple resource datasets. Chinese J. Atmos. Sci., 47 , 837–852. (in Chinese)
    Liu, G., G. L. Feng, Y. L. Qin, et al., 2017: “Cumulative effect” of cold vortex precipitation in Northeast China in early summer. Chinese J. Atmos. Sci., 41, 202–212, https://doi.org/10.3878/j.issn.1006-9895.1602.15231. (in Chinese)
    Miao, C. S., Z. W. Wu, J. H. He, et al., 2006a: The anomalous features of the Northeast Cold Vortex during the first flood period in the last 50 years and its correlation with rainfall in South China. Chinese J. Atmos. Sci., 30, 1249–1256, https://doi.org/10.3878/j.issn.1006-9895.2006.06.19. (in Chinese)
    Miao, C. S., Z. W. Wu, and J. H. He, 2006b: Relationship among the northern hemisphere annual mode, the Northeast China Cold Vortex and precipitation during the first yearly rainy period in South China. J. Trop. Meteor., 22, 593–599, https://doi.org/10.3969/j.issn.1004-4965.2006.06.011. (in Chinese)
    Nieto, R., L. Gimeno, L. de la Torre, et al., 2005: Climatological features of cutoff low systems in the northern hemisphere. J. Climate, 18, 3085–3103, https://doi.org/10.1175/JCLI3386.1.
    Niu, Z. Y., X. L. Zou, and D. Q. Li, 2021: Northeast China Cold Vortex observed by FY-3 MWTS-2 and MetOp AMSU-A. J. Geophys. Res. Atmos., 126, e2021JD035471, https://doi.org/10.1029/2021JD035471.
    Price, J. D., and G. Vaughan, 1992: Statistical studies of cut-off-low systems. Ann. Geophys., 10, 96–102.
    Ren, S. L., X. Fang, N. M. Lu, et al., 2019: Recognition method of the Tibetan plateau vortex based on meteorological satellite data. J. Appl. Meteor. Sci., 30, 345–359, https://doi.org/10.11898/1001-7313.20190308. (in Chinese)
    Ren, S. L., J. Y. Jiang, X. Fang, et al., 2022: FY-4A/GIIRS temperature validation in winter and application to cold wave monitoring. J. Meteor. Res., 36, 658–676, https://doi.org/10.1007/s13351-022-2015-4.
    Ren, S. L., D. Y. Qin, N. Niu, et al., 2023: South Asian High identification and rainstorm monitoring using Fengyun-4-derived atmospheric motion vectors. Atmosphere, 14, 1606, https://doi.org/10.3390/atmos14111606.
    Shi, Y. Q., S. Y. Gao, S. Jing, et al., 2022: Analysis of the characteristics and mechanism of a regional heavy rain event in Liaoning Province. Plateau Meteor., 41, 630–645. (in Chinese)
    Sun, L., X. Y. Zheng, and Q. Wang, 1994: The climatological characteristics of Northeast Cold Vortex in China. J. Appl. Meteor. Sci., 5, 297–303. (in Chinese)
    Wang, C. W., H. M. Xu, L. Ren, et al., 2012: The objective identification method of Northeast Cold Vortex. J. Meteor. Environ., 28, 1–4, https://doi.org/10.3969/j.issn.1673-503X.2012.02.001. (in Chinese)
    Wang, W. Z., J. Li, C. L. Hu, et al., 2017: A review of definition, identification and quantitative investigation on Northeast Cold Vortex. Scientia Meteor. Sinica, 37, 394–402, https://doi.org/10.3969/2015jms.0092. (in Chinese)
    Wang, Z. M., J. B. Li, F. X. Wang, et al., 2015: Asymmetric characteristics of the Northeast Cold Vortex and its effect on heavy rain. Plateau Meteor., 34, 1721–1731. (in Chinese)
    Wu, D., S. W. Shou, and X. P. Yao, 2010: Dry intrusion’s feature and its relationship with rainfall regions on cold vortex heavy rain process in Northeast China. Torrential Rain and Disasters, 29, 111–116, https://doi.org/10.3969/j.issn.1004-9045.2010.02.002. (in Chinese)
    Wu, Y. Q., Q. Yan, X. Y. Kang, et al., 2009: Anomaly characteristics of persistent activity Northeast Cold Vortex in May of 2008. J. Meteor. Environ., 25, 13–17, https://doi.org/10.3969/j.issn.1673-503X.2009.05.003. (in Chinese)
    Yin, S., J. Feng, and J. P. Li, 2013: Influences of the preceding winter northern hemisphere annular mode on the spring extreme low temperature events in the north of eastern China. Acta Meteor. Sinica, 71, 96–108, https://doi.org/10.11676/qxxb2013.008. (in Chinese)
    Zhang, C., Y. Q. Wang, X. Y. Shen, et al., 2019: Theoretical analysis and numerical study on the development mechanism of squall line in the Northeast Cold Vortex. Chinese J. Atmos. Sci., 43, 361–371, https://doi.org/10.3878/j.issn.1006-9895.1806.18101. (in Chinese)
    Zhang, F. Q., 2001: Computer recognition of Northeast Cold Vortex. Meteor. Mon., 27, 46–49. (in Chinese)
    Zhang, G. H., and C. Z. Liu, 2010: The objective recognition of Northeast Cold Vortex and its monitoring method. Proceedings of the 27th Annual Meeting of the Chinese Meteorological Society on Disaster Weather Research and Forecast, Beijing, 21–23 October, Chinese Meteorological Society. (in Chinese)
    Zhang, J. G., S. H. Zhao, Y. Liu, et al., 2021: Observations and analysis of vertical structure of cloud and precipitation caused by a cold vortex in Northeast China. J. Meteor. Environ., 37, 1–8, https://doi.org/10.3969/j.issn.1673-503X.2021.04.001. (in Chinese)
    Zhu, Y. X., 2023: Statistical Methods in the Atmospheric Sciences, China Meteorological Press, Beijing, 685 pp. (in Chinese)
  • Related Articles

  • Other Related Supplements

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return