1. Introduction
Clear-air turbulence (CAT) can induce bumpiness of aircraft at high altitudes in free atmosphere without convective clouds (excluding nonconvective cirrus) (Chambers 1955). It occurs without advanced warning and is primarily concentrated at cruising altitudes, typically over 6.0 km (FL200) above mean sea level (MSL). CAT has become one of the major factors affecting flight safety and economic benefits of civil aviation (Sharman et al. 2012). It has been getting more and more urgent to improve the accuracy of CAT prediction because of the continuous increase in global flight volume (Sharman and Lane 2016). In recent years, numerical simulation technologies have been used to investigate CAT generation mechanisms in order to improve the forecasting accuracy and to minimize the impacts of CAT events on pilots and airlines (e.g., Trier and Sharman 2009; Trier et al. 2010, 2012, 2022; Kim and Chun 2010, 2012; Kim et al. 2014, 2015; Huang et al. 2015; Park et al. 2016; Lee and Chun 2018).
Previous studies indicated three main factors affecting CAT generation: vertical wind shear (VWS) associated with upper-level front–jet system, gravity wave caused by various sources, and inertial instability related with strong anticyclonic flows (Kim and Chun 2010; Sharman et al. 2012; Lee and Chun 2018; Storer et al. 2019). Among these three factors, VWS is the most predominant one (Park et al. 2016), which frequently results in Kelvin–Helmholtz (K–H) instability (Dutton and Panofsky 1970). When the unstable atmosphere develops into a critical condition, large-scale fluctuations will break down and overturn, with turbulent kinetic energy drawn from average flows and eventually transferred to small-scale turbulent eddies (Sharman et al. 2006; Lane et al. 2012). A local Richardson number (Ri) less than 0.25 and a local Ri less than 1 can be used to identify K–H instability in a stable atmospheric stratification (Miles 1961) and a more realistic three-dimensional nonlinear atmosphere (Miles 1986; Sharman et al. 2012; Lee et al. 2023), respectively. Moreover, CAT events caused by K–H instability typically occur above or below upper-level jet streams, the entrance region of jet streams, and the intersection region of two distinct flows where strong VWS frequently occurs (Roach 1970; Kaplan et al. 2005).
Gravity wave is another important source of CAT generation. It can reduce local Ri and cause K–H instability in an environment that would not normally generate turbulence (Storer et al. 2019). Williams et al. (2003, 2005, 2008) summarized various sources of gravity wave generation, including complex terrain and convective system. The amplitude of vertically propagating mountain wave increases with height because of the decrease in air density, and then the wave may break at a higher altitude and induce CAT (Doyle et al. 2005). Atmospheric stability increasing near the tropopause and stationary mountain waves approaching a critical level (background wind speed equals zero) can also increase the potential of wave breaking (Kim and Chun 2010). The CAT induced by convective gravity waves may depend on a large VWS above or around the developing convective cloud tops (Lane et al. 2003, 2004; Koch et al. 2005; Lane and Sharman 2008). The inertial instability associated with anticyclonic flows is also a factor influencing CAT generation. According to Knox (1997), inertial instability and/or geostrophic adjustment is thought to be a more plausible source of CAT generation in a strong anticyclonic flow, because the deformation caused by upper-level frontogenesis is insufficient in anticyclonic conditions.
To study the possible generation mechanisms of CAT, Kim and Chun (2010) divided nine moderate-or-greater (MOG) CAT events over South Korea on 2 April 2007 into three groups according to their locations. High-resolution numerical simulations indicated that strong VWS in upper-level front–jet systems induces exacerbation of inertial instability near Jeju Island and thus generates CATs on the anticyclonic side of the jet stream. The CAT events in the eastern mountainous regions are caused by mountain wave breaking. Kim et al. (2014) studied a series of MOG CAT events occurring inside and above the ribbon cirrus along the flight route from Tokyo to Hawaii on 9 September 2010 with high-resolution numerical simulations. The results showed that the strong VWS associated with convections leads to deformation and mixing of flows near the cloud top, which ultimately causes localized CAT events. Lee and Chun (2018) studied a CAT event occurring over the Yellow Sea on 13 February 2013. They indicated that during the northward movement of a jet stream, an upper-level front on its northern edge intersects with a trough over Northeast China, which induces strong VWS and tropopause folding and thus generates a CAT.
On 13 November 2019, the in situ aircraft automatic observation equipment of China Eastern Airlines (CEA) recorded nine severe-or-greater (SOG) CAT events over central and eastern China between 0000 and 1200 UTC. It is the day with the most SOG CAT events in the winter of 2019 in the above region. In this study, high-resolution simulations are carried out to reproduce these CAT events and investigate their generation mechanisms. The description of these events and the experiment design are presented in section 2. The simulation performance is evaluated in section 3, and the generation mechanisms at different scales are investigated in section 4. Section 5 gives the summary and discussion.
2. Data and method
a. Description of the CAT events on 13 November 2019
To monitor aviation turbulence, CEA has equipped its commercial aircraft with autonomous equipment to detect eddy dissipation rate (EDR; m2/3 s−1) since late 2018. This dataset has compensated for the lack of reliable turbulence observations in China (Hu et al. 2022). The in situ aircraft observations are downlinked every 20 min during each flight, with the EDR greater than 0.1 m2/3 s−1 captured. Each record contains the following information: flight number, airports of departure and destination, observation location (latitude, longitude, and flight altitude), observation time, temperature, wind speed, wind direction, and EDR intensity. On 13 November 2019, seven CEA aircraft encountered nine SOG (ICAO 2001; EDR greater than 0.5 m2/3 s−1) turbulence events from 0000 (0800) to 1200 (2000) UTC (LST) over central and eastern China at or above 6 km MSL.
The detailed information of the nine turbulence events are listed in Table 1, and their locations are shown in Fig. 1. All the turbulence events occurred below the tropopause at altitudes between 6.0 and 6.7 km and were concentrated during four periods. Events 1–4 were recorded over Taihang Mountains and North China Plain in the morning (0013–0035 UTC), and events 1–3 were observed by the same aircraft. At midday (0618 UTC), event 5 occurred over the southern Shandong Peninsula. Events 6 and 7 occurred in the evening (0933–0934 UTC) over the Yellow Sea and northern Qinling Mountains, respectively. At nighttime (1101–1108 UTC), events 8 and 9 were distributed over the south of North China Plain. The above events were all collected by medium-sized aircraft, such as Airbus (A) 320 and A321, and Boeing (B) 737. Therefore, the uncertainty of EDR intensity caused by different aircraft types can be ignored.
Information about nine SOG turbulence events over central and eastern China on 13 Nov 2019, including time, longitude, latitude, flight altitude, EDR intensity, flight number, and aircraft type.
Locations of nine SOG turbulence events (crosses) superimposed on terrain height (m; shading) on 13 Nov 2019 over central and eastern China. The numbers are corresponding to those listed in Table 1. The blue dots represent the radiosonde stations.
Citation: Weather and Forecasting 38, 9; 10.1175/WAF-D-23-0015.1
Figure 2 shows the cloud-top information for the nine turbulence events calculated from the Advanced Himawari Imager (AHI) data (Zhuge et al. 2021). On each AHI pixel, the cloud-top phase is classified into 10 types, namely, clear, probably clear, probably cloudy, water, supercooled, mixed, thick ice, cirrus, overlap, and overshooting. Between 0000 and 0600 UTC, although clouds are observed over most regions, turbulence events 1–5 occur in clear or probably clear conditions (Figs. 2a,b). The clouds gradually move southward from 0900 to 1100 UTC, and turbulence events 6–9 all occur in clear conditions (Figs. 2c,d). The above analysis indicates that all the turbulence events in this study are CAT events.
Spatial distributions of cloud-top phase at (a) 0000, (b) 0600, (c) 0900, and (d) 1100 UTC 13 Nov 2019. The locations of CAT events are depicted by crosses with numbers corresponding to those listed in Table 1.
Citation: Weather and Forecasting 38, 9; 10.1175/WAF-D-23-0015.1
b. Numerical experiment design
The new generation nonhydrostatic mesoscale model WRF (Weather Research and Forecasting) is used in this study, which was jointly developed by the National Center for Atmospheric Research (NCAR) and the National Oceanic and Atmospheric Administration (NOAA) (Skamarock et al. 2008). It has been widely used in numerical simulation studies of CAT (e.g., Kim and Chun 2010; Kim et al. 2014; Park et al. 2016; Lee and Chun 2018).
In this study, the numerical experiment is conducted over two domains with a horizontal grid resolution of 5 and 1 km, respectively (Fig. 3). Domain 1 (D1) covers central and eastern China and domain 2 (D2) focuses on the area where the nine SOG CAT events occurred. The experiment adopts 88 vertical levels with a vertical resolution of 300 m and the top at 20 hPa. The initial and boundary conditions are provided by the ERA5 reanalysis data from the European Centre for Medium-Range Weather Forecasts (ECMWF) (Hersbach et al. 2020). It has a horizontal resolution of 0.25° × 0.25° and a temporal resolution of 6 h. The experimental integration lasts 24 h in D1 (1200 UTC 12 November–1200 UTC 13 November) and 18 h in D2 (1800 UTC 12 November–1200 UTC 13 November), and the atmospheric variables are obtained at 5-min intervals. The parameterization schemes in the experiment include the new Thompson microphysics scheme (Thompson et al. 2008), the RRTMG scheme for longwave and shortwave radiation (Iacono et al. 2008), the Mellor–Yamada–Janjić planetary boundary layer scheme (Mellor and Yamada 1982; Janjić 1994, 2002), the ETA similarity surface layer scheme (Janjić 1996, 2002), and the 5-layer thermal diffusion land surface scheme (Dudhia 1996).
Outer and inner domains of the WRF experiment with a horizontal resolution of 5 km (D1, gray shading) and 1 km (D2, color shading), respectively.
Citation: Weather and Forecasting 38, 9; 10.1175/WAF-D-23-0015.1
The Integrated Global Radiosonde Archive (IGRA) data at eight radiosonde stations are used to evaluate the numerical simulation results, and the station locations are shown in Fig. 1. The IGRA data are provided by the National Climate Data Center (NCDC) (ftp://ftp.ncdc.noaa.gov/), and it is composed of observations from radiosondes and pilot balloons at more than 2800 observational stations around the world. The variables of IGRA data include pressure, temperature, geopotential height, relative humidity, wind speed, and wind direction.
c. CAT diagnostics
Following previous studies (Hu et al. 2022; Lee et al. 2023), we employ seven CAT diagnostics (Table 2) to investigate CAT generation mechanisms in this study. These diagnostics can be divided into two groups. TI1 and TI2, as well as their components VWS and DEF, are empirical diagnostics derived from statistical and regression analysis based on turbulence observations (Sharman et al. 2006; Ellrod and Knapp 1992). The terms N2, Ri, and PV are theoretical diagnostics derived from fundamental physical quantities associated with atmospheric motions, which are directly related to convective instability (N2 less than 0), K–H instability (Ri less than 1), and inertial instability (PV less than 0), respectively (Lee et al. 2023).
Brief definitions and algorithms of seven CAT diagnostics. The terms θ, g, and T represent potential temperature, gravity acceleration, and absolute temperature, respectively; u and υ are zonal and meridional wind components, respectively; DSH, DST, and DIV represent shearing deformation [DSH = (∂υ/∂x) + (∂u/∂y)], stretching deformation [DST = (∂u/∂x) + (∂υ/∂y)], and horizontal divergence [DIV = (∂u/∂x) + (∂υ/∂y)], respectively; 1 PVU equals 10−6 K kg−1 m2 s−1.
3. Model evaluation
a. Synoptic-scale flow in D1
The synoptic-scale flow in D1 is evaluated against the ERA5 data, including sea level pressure, 300-hPa geopotential height, and wind field at 0000 UTC 13 November 2019 (Fig. 4). The sea level pressure demonstrates a dipole pattern with high and low pressure centers over Hetao Plain and northeast Korean Peninsula, respectively, and a large pressure gradient is found between 110° and 120°E (Figs. 4a,b). This distributions is consistent with the climatological mean pattern of the East Asian winter monsoon (EAWM), which is characterized by the Siberian high and Aleutian low (Jhun and Lee 2004; Chang and Lu 2012). The Siberian high is generally located in the region of 40°–60°N and 80°–120°E (Jhun and Lee 2004). As it intensifies and leaves the source, strong north and northeast winds are produced on its east and south edges, respectively. The WRF simulation shifts the continental high pressure center southward and eastward, which causes the isobaric distribution on the southeast side of the high pressure to be different from the reanalysis. Moreover, the simulation underestimates (overestimates) the high (low) pressure (Fig. 4b). The 300-hPa jet stream is located between 35° and 44°N and inclines from northeast to southwest, with a maximum horizontal wind speed of 79.7 m s−1. Additionally, there is a curved low pressure trough on the cyclonic side of the jet stream entrance zone (Fig. 4c). The WRF Model accurately captures the jet stream location and produces the maximum horizontal wind speed (74.7 m s−1) (Fig. 4d).
Spatial distributions of (a),(b) sea level pressure (hPa; shading) and (c),(d) horizontal wind speed (m s−1; shading) superimposed on geopotential height (6-dagpm interval; contours) at 300 hPa obtained from (left) ERA5 data and (right) WRF simulation in D1 at 0000 UTC 13 Nov 2019.
Citation: Weather and Forecasting 38, 9; 10.1175/WAF-D-23-0015.1
b. Atmospheric vertical structure in D2
Since the occurrence altitude of the CAT events spreads between 6.0 and 6.7 km, the capacity of the WRF Model in reproducing the vertical structure of the atmosphere is also crucial for generation mechanism analysis. Figure 5 shows the vertical profiles of temperature, dewpoint temperature, and horizontal wind speed obtained from IGRA data and WRF simulation in D2 at 0000 UTC 13 November 2019. The temperature decreases with altitude at almost all the radiosonde stations except Xingtai station with an inversion structure over 150 hPa (Fig. 5a). WRF well reproduces the vertical structure of temperature, with the simulated profiles closely matching the observations. The observed vertical profile of dewpoint temperature demonstrates a dry tongue at 700 hPa at Xingtai and Zhengzhou stations; however, WRF simulates the dry tongue near 925 hPa (Figs. 5a,d). In addition, WRF simulates a dry tongue of dewpoint temperature near 600 hPa at Xuzhou station, which is absent in the observation (Fig. 5e). At Zhangqiu and Sheyang stations, the model produces a wetter layer than the observations in the middle troposphere (Figs. 5b,f). The vertical profile of wind speed demonstrates a peak at around 200 hPa, which reflects the upper-level jet stream, at all the stations except Xingtai. The WRF Model is capable of simulating the wind field profiles at most stations. However, the WRF Model underestimates the wind speed at Xingtai station and falsely produces a wind disturbance in the lower troposphere at Xingtai, Zhengzhou, Xuzhou, and Nanyang stations (Figs. 5a,d,e,g). In general, the WRF Model well reproduces the vertical distribution characteristics of temperature, dewpoint temperature, and wind field in this numerical experiment.
Vertical profiles of temperature (°C; red line), dewpoint temperature (°C; green line), and horizontal wind speed (m s−1; black line) obtained from IGRA data (solid line) and WRF simulation (dotted line) at eight IGRA stations at 0000 UTC 13 Nov 2019.
Citation: Weather and Forecasting 38, 9; 10.1175/WAF-D-23-0015.1
4. Generation mechanism
a. Synoptic-scale analysis in D1
In this section, the synoptic-scale characteristics of atmospheric circulation corresponding to the CAT events are investigated. Figure 6 shows the spatial distributions of horizontal wind speed, horizontal wind direction, and potential temperature obtained from WRF simulation in D1 at 0025, 0620, 0935, and 1105 UTC 13 November 2019 corresponding to CAT events 1–4, 5, 6–7, and 8–9, respectively. The analysis altitude (6.4 km) is the average altitude of the nine events. At 0025 UTC, a strong meridional potential temperature gradient accompanied by an upper-level front–jet system spreads from the north of Shandong Peninsula to the south of Hetao Plain. CAT events 1–4 occur in the convergence zone of northwest wind and southwest wind in the entrance region of the jet stream (Fig. 6a). From 0025 to 0620 UTC, the maximum horizontal wind speed of the jet stream increases from 59.0 to 63.0 m s−1, and CAT event 5 is generated in the strong potential temperature gradient zone on the cyclonic side of the jet stream (Fig. 6b). The front–jet system keeps moving southward accompanied with the weakening of the meridional potential temperature gradient. At 0935 and 1105 UTC, the maximum horizontal wind speed of the jet stream is 67.3 and 66.8 m s−1, respectively, and CAT events 6–7 and 8–9 occur on the cyclonic side of the jet stream (Figs. 6c,d). Previous studies have also shown similar atmospheric background for strong CAT events near front–jet systems in East Asia (e.g., Kim and Chun 2010; Lee and Chun 2018).
Spatial distributions of horizontal wind speed (m s−1; shading) superimposed on horizontal wind direction (arrowhead) and potential temperature (4-K intervals; contours) obtained from the WRF simulation in D1 at 6.4 km at (a) 0025, (b) 0620, (c) 0935, and (d) 1105 UTC 13 Nov 2019. The locations of the CAT events are depicted by crosses with numbers corresponding to those listed in Table 1.
Citation: Weather and Forecasting 38, 9; 10.1175/WAF-D-23-0015.1
Figure 7 shows the spatial distribution of relative humidity superimposed on horizontal wind speed at four moments. At 0025 UTC, a wet tongue is located on the anticyclonic side of the jet stream with a “northeast–southwest” distribution. Then, the front–jet system moves southward from 0620 to 1105 UTC, and the wet tongue turns into an “east–west” distribution. The nine CAT events are concentrated in the dry environment on the north side of the wet tongue, which is also confirmed by the satellite cloud product shown in Fig. 2.
As in Fig. 6, but for spatial distributions of relative humidity (%; shading) superimposed on horizontal wind speed (5 m s−1 intervals; contours).
Citation: Weather and Forecasting 38, 9; 10.1175/WAF-D-23-0015.1
The vertical characteristics of the synoptic-scale circulation are further investigated to understand the relationship between the CAT events and the upper-level front–jet system. Figure 8 shows the vertical cross sections of zonal horizontal wind speed superimposed on potential temperature and 1.5-PVU (1 PVU = 10−6 K kg−1 m2 s−1) line along 112°E (events 4, 7), 116°E (events 1–3, 5, 8–9), and 120°E (event 6) obtained from WRF simulation in D1. At 1800 UTC 12 November 2019, a weak upper-level jet stream is located at 12 km along 112°E. Meanwhile, the contour lines of potential temperature are inclined and dense in the area of 34°–38°N, 4–8 km below the jet stream, which indicates a frontal zone (Fig. 8a). The tropopause height (1.5 PVU) decreases from 14 km on the anticyclonic side of the jet stream to 8 km on the cyclonic side (Figs. 8a–c). At 0025 UTC 13 November 2019, the jet stream gets stronger, and the maximum zonal horizontal wind speed reaches above 75 m s−1. CAT event 4 occurs at the boundary of the high wind speed zone, which extends along the cyclonic side of the jet stream (Fig. 8d). At 0620 UTC, the front–jet system gets more southward, with the tropopause folding reaching around 33°N, and the upper troposphere exhibits high stability (Fig. 8g). At 0935 and 1105 UTC, the tropopause folding is no longer distinct due to the weakening of the front–jet system, and CAT event 7 occurs on the cyclonic side of the jet stream at 0935 UTC (Figs. 8j,m).
Vertical cross sections of zonal horizontal wind speed (m s−1; shading) superimposed on potential temperature (4-K intervals; thin line) and 1.5-PVU line (thick line) obtained from the WRF simulation in D1 along L1–L3 shown in Fig. 6b at (a)–(c) 1800 UTC 12 Nov 2019 and (d)–(f) 0025, (g)–(i) 0620, (j)–(l) 0935, and (m)–(o) 1105 UTC 13 Nov 2019. The locations of the CAT events are depicted by crosses with numbers corresponding to those listed in Table 1.
Citation: Weather and Forecasting 38, 9; 10.1175/WAF-D-23-0015.1
On the 116°E cross section, the contour lines of potential temperature tend to get denser from 1800 UTC 12 November to 0025 UTC 13 November, indicating an increase in frontal slope, and CAT events 1–3 occur in the frontal zone (Figs. 8b,e). The zonal horizontal wind speed of the jet stream reaches the highest at 0620 UTC when CAT event 5 occurs (Fig. 8h). From 0025 to 0935 UTC, the tropopause folding is fully evident (Figs. 8e,h,k), and CAT events 8–9 occur in the frontal zone at 1105 UTC (Fig. 8n). It is found that the descending of the high wind speed zone is well consistent with the strengthening of the potential temperature gradient below the tropopause folding. This is similar to the findings in previous studies (e.g., Koch et al. 2005; Kim and Chun 2010; Lee and Chun 2018). Between 0620 and 1105 UTC, the tropopause folding is more evident along 120°E than along 116°E (Figs. 8i,l,o), and CAT event 6 occurs at 0935 UTC along 120°E. In a word, the SOG CAT events in this study are well correlated with the development of the upper-level front–jet system and tropopause folding, and the downward extension of the 1.5-PVU line to the lower troposphere is favorable for the intensification of the jet stream.
b. CAT diagnostics analysis in D2
Figure 9 depicts the spatial distributions of seven CAT diagnostics obtained from the WRF simulation in D2 at four moments on 13 November 2019. Results show that both N2 and PV are evidently greater than 0 on the cyclonic side of the upper-level jet stream (Figs. 9a,g), indicating a strong atmospheric stability. That is to say, the CAT events in this study are not directly correlated with convective and inertial instability. In addition, a strong VWS effect is presented on the cyclonic side of the jet stream. The VWS gradually strengthens from 0025 to 0620 UTC and then gradually weakens from 0935 to 1105 UTC (Fig. 9b). This process is consistent with the development of the front–jet system (Fig. 8). The increase and decrease in meridional gradient of potential temperature leads to strengthening and weakening of the jet stream, respectively, through a thermal wind response, and thus induces increase and decrease in VWS, respectively (Storer et al. 2019). DEF, TI1, and TI2 experience a similar evolution process with the above diagnostics (Figs. 9d–f), and Ri demonstrates an opposite phase to the other six diagnostics (Fig. 9c). The local Ri less than 1 indicates K–H instability in a three-dimensional nonlinear atmosphere when VWS is large and/or N2 is small (Miles 1986; Sharman et al. 2012; Lee et al. 2023). In this study, all the CAT events occur in the regions with large VWS and small Ri. Herein, it could be concluded that K–H instability plays an important role in CAT generation, with smaller Ri induced by large VWS (denominator) rather than weak stratification (numerator, N2).
Spatial distributions of CAT diagnostics of (a) N2, (b) VWS, (c) Ri, (d) DEF, (e) TI1, (f) TI2, and (g) DEF superimposed on horizontal wind speed (5 m s−1 intervals; contours) obtained from the WRF simulation in D2 at 6.4 km at 0025, 0620, 0935, and 1105 UTC 13 Nov 2019. The locations of the CAT events are depicted by crosses with numbers corresponding to those listed in Table 1.
Citation: Weather and Forecasting 38, 9; 10.1175/WAF-D-23-0015.1
Figure 10 further shows the vertical cross sections of six CAT diagnostics, with potential temperature and 1.5-PVU line superimposed. TI2 is not chosen in this section, because its diagnostic performance is highly similar to TI1. The results suggest a considerable stability (large N2 and PV) above the tropopause (1.5 PVU). In addition, N2 and PV are large at altitudes between 4 and 8 km, which is unfavorable for CAT generation (Figs. 10a,f). It is clear that all the CAT events occur in the regions with relatively large VWS and small Ri (less than 1), which are inclined along the dense contour lines of potential temperature (Figs. 10b,c). Meanwhile, VWS and Ri demonstrate more fluctuations along 112°E than on the other two cross sections, which may be affected by the upward transmission of mountain waves. The occurrence of CAT event shows a weak correlation with the vertical distribution of DEF (Fig. 10d); however, the vertical pattern of TI1 exhibits a good consistency with the CAT events (Fig. 10e).
Vertical cross sections of CAT diagnostics of (a) N2, (b) VWS, (c) Ri, (d) DEF, (e) TI1, and (f) DEF superimposed on potential temperature (4-K intervals; thin line) and 1.5-PVU line (thick line) obtained from the WRF simulation in D2 over the region between the two blue dashed lines shown in Fig. 8o. The locations of the CAT events are depicted by crosses with numbers corresponding to those listed in Table 1.
Citation: Weather and Forecasting 38, 9; 10.1175/WAF-D-23-0015.1
To investigate the spatial resolution dependency of the simulated CAT diagnostics, the VWS and Ri for the nine CAT events are calculated from the simulations over both D1 and D2 (Table 3). These two diagnostics are the key factors affecting CAT generation. Over D1, the VWS ranges from 1.55 × 10−2 (event 7) to 2.51 × 10−2 s−1 (event 6), and the local Ri ranges from 0.43 (event 6) to 0.85 (event 1). The CAT diagnostics values over D2 are quite similar to those over D1 (from 1.60 × 10−2 to 2.53 × 10−2 s−1 for VWS and from 0.42 to 0.84 for Ri), implying little impact of model resolution on simulation results. The correlation coefficients (CORRs) between the observed EDR intensity and the simulated CAT diagnostics are not significant in both simulations, which might be due to the small size of the CAT events. Nevertheless, these diagnostics have shown outstanding performances in diagnosing CAT potential in China based on thousands of CAT events (Hu et al. 2022).
Observed EDR intensity and WRF simulated VWS and Ri in D1 and D2 for the nine SOG CAT events over central and eastern China on 13 Nov 2019. The numbers correspond to those listed in Table 1.
c. Mountain-wave effects
The vertical motion induced by mountain waves may contribute to the CAT generation, so the vertical velocity associated with the nine CAT events is analyzed (Fig. 11). The WRF Model successfully reproduces an evident small-scale wave pattern over the northwest part of the domain, where CAT event 4 occurs at 0025 UTC (Fig. 11a). In addition, CAT event 7 is also associated with a weak small-scale wave system at 0935 UTC (Fig. 11c). The generation of these mountain waves is related to the complex terrain in the domain, such as Loess Plateau, Taihang Mountains, and Qinling Mountains. Mountain waves have also been captured during the strong CAT events along the east coasts of the Korean Peninsula and over the Yellow Sea (Kim and Chun 2010; Lee and Chun 2018). The other seven CAT events are less influenced by the terrain, and there is no obvious vertical motion around them (Figs. 11b,d).
Spatial distributions of vertical velocity (m s−1; shading) superimposed on horizontal wind direction (arrowhead) at 6.4 km obtained from the WRF simulation in D1 at (a) 0025, (b) 0620, (c) 0935, and (d) 1105 UTC 13 Nov 2019. The locations of the CAT events are depicted by crosses with numbers corresponding to those listed in Table 1.
Citation: Weather and Forecasting 38, 9; 10.1175/WAF-D-23-0015.1
To better understand the vertical structure of the small-scale wave, the vertical velocity, and Ri along the cross sections of 35.9° and 34.3°N in D2 are shown in Fig. 12. It can be found that CAT event 4 occurs in the region with Ri less than 1, which is distributed in a belt and concentrated between 6 and 8 km. The vertical velocity fluctuation extends from the surface to the altitude above 15 km, particularly over the region with a significant topographic gradient (Fig. 12a). Similar to CAT event 4, CAT event 7 also occurs in a band with Ri less than 1. However, the small-scale waves are only concentrated below 5 km around the location of CAT event 7 (Fig. 12b).
Vertical cross sections of vertical velocity (m s−1; red and blue shading) and Ri (green shading) superimposed on potential temperature (4-K intervals; thin line) and 1.5-PVU line (thick line) obtained from the WRF simulation in D2 at (a) 0025 and (b) 0935 UTC 13 Nov 2019. The locations of the CAT events are depicted by crosses with the numbers corresponding to those listed in Table 1.
Citation: Weather and Forecasting 38, 9; 10.1175/WAF-D-23-0015.1
d. Typical flight process
The CEA aircraft continuously downlink observation data when the EDR greater than 0.1 m2/3 s−1, and typical flight processes with frequent turbulence events can be selected to analyze the generation mechanism of strong CAT. Figure 13 displays the flight track of MU2429 from Jinan Yaoqiang International Airport to Xi’an Xianyang International Airport on 13 November 2019 (Fig. 13a), along with the flight altitude, wind direction, and wind speed during the flight process (Fig. 13b). The aircraft quickly climbs to an altitude of 7.1 km after takeoff and continuously encounters light (0.1 m2/3 s−1 < EDR ≤ 0.3 m2/3 s−1) and moderate (0.3 m2/3 s−1 < EDR ≤ 0.5 m2/3 s−1) CAT. It is the first stage (S1) of the selected typical flight process (between 0007 and 0017 UTC). To ensure flight safety, the flight altitude is reduced to 5.4 km immediately. During the second stage of the typical flight process (S2) between 0018 and 0030 UTC, only one CAT event occurs. Then, the aircraft rises to 6.0 km at around 0031 UTC and encounters CATs again (third stage of the typical flight process, S3). After 0042 UTC, the flight is relatively stable, and only a few light CAT events are encountered before landing (fourth stage of the typical flight process, S4). A westerly wind flow prevails during the whole flight process, and the wind direction temporarily changes during S1 when severe CAT events occurs (Fig. 13b). Moreover, the wind speed rapidly increases before S1 and during S2 (Fig. 13b).
(a) Flight track (black line) superimposed on terrain height (km; shading) and CAT events (color crosses), and (b) flight altitude (km; blue line), horizontal wind speed (m s−1; red line), and horizontal wind direction (°; black line) superimposed on EDR intensity (m2/3 s−1; color shading) obtained from the in situ aircraft observations of MU2429 on 13 Nov 2019. The red, yellow, and green crosses and shading represent severe, moderate, and light CAT events, respectively.
Citation: Weather and Forecasting 38, 9; 10.1175/WAF-D-23-0015.1
Figure 14 shows the zonal–temporal and vertical–temporal variations of VWS and Ri obtained from the WRF simulation in D1 during the three stages of the typical flight. The zonal–temporal and vertical–temporal cross sections are selected based on the vertical–meridional and zonal–meridional mean of the diagnostics, respectively. The results demonstrate that all the CAT events occur in an environment with VWS greater than 1.3 × 10−2 s−1 and local Ri less than 1 during S1 (Figs. 14a,b). During S2, the aircraft descends to approximately 5.4 km with low VWS and high Ri, and only one CAT event is observed (Figs. 14c,d). When the flight altitude reaches 6.0 km (S3), favorable atmospheric conditions trigger continuous light CAT events (Figs. 14e,f). It is noted that the consistency between the location of CAT event and the region with large VWS and/or small Ri during S3 is not as high as that during S1, which might be due to the relatively weaker intensity of the CAT events.
Zonal–temporal (6.6 km, 36.7°–37.0°N mean for S1; 5.4 km, 36.2°–36.6°N mean for S2; 6.0 km, 35.7°–36.1°N mean for S3) and vertical–temporal (115.6°–116.5°E, 36.7°–37.0°N mean for S1; 114.3°–115.5°E, 36.2°–36.6°N mean for S2; and 113.4°–114.2°E, 35.7°–36.1°N mean for S3) distributions of VWS (10−2 s−1; shading) and Ri (contours) obtained from the WRF simulation in D1 during (a),(b) S1; (c),(d) S2; and (e),(f) S3 of the typical flight process on 13 Nov 2019. The thick black lines and the color crosses represent flight track and CAT events (red for severe, yellow for moderate, and green for light CAT), respectively. The regions surrounded by horizontal and vertical blue dashed lines represents the flight ranges in each flight stage.
Citation: Weather and Forecasting 38, 9; 10.1175/WAF-D-23-0015.1
5. Summary and discussion
On 13 November 2019, CEA aircraft observed nine SOG CAT events over central and eastern China between 0000 and 1200 UTC. These events were concentrated in four periods and mainly occurred at altitudes between 6.0 and 6.7 km. A nested numerical simulation is conducted with the WRF Model at horizontal resolutions of 5 and 1 km and seven CAT diagnostics are calculated to investigate possible generation mechanisms of the CAT events.
The WRF Model well reproduces the spatial distributions of synoptic-scale sea level pressure, geopotential height, and wind field over central and eastern China, while the simulated high pressure center deviates from the observation in some degree. The vertical characteristics of temperature, dewpoint temperature, and wind field are also well captured when compared with radiosonde data, although there are some biases in the lower troposphere. On 13 November, an enhanced front–jet system is observed over central and eastern China with its maximum wind speed increasing from 59.0 m s−1 in the day to 67.3 m s−1 at night. An obvious tropopause folding occurs under the upper-level front–jet system, and the tropopause height decreases from 14 km on the anticyclonic side of the jet stream to 8 km on the cyclonic side. All the studied CAT events occur in regions with a significant meridional potential temperature gradient.
The diagnostic analysis indicates that N2 and PV are both evidently greater than 0 when CAT events occur, proving that CAT is not directly induced by convective and inertial instability. All the CAT events in this study occur in the regions with large VWS (1.55 × 10−2–2.53 × 10−2 s−1) and small Ri (0.42–0.85), that is to say, K–H instability indicated by local Ri less than 1 is crucial to CAT generation. Additionally, small-scale waves induced by complex terrain are captured around some of the events. However, the majority of the waves are concentrated below the CAT event altitude. The investigation of a typical flight process with continuous CAT events on 13 November 2019 also shows that large VWS (greater than 1.3 × 10−2 s−1) accompanied with small Ri (less than 1) is favorable for CAT generation. It is noteworthy that both VWS and Ri have a better diagnostic effect for stronger CAT events.
The numerical experiment in this study is helpful for understanding the generation mechanism of CAT over central and eastern China and improving the accuracy of CAT prediction. It confirms the important role of K–H instability in CAT generation, and more simulation studies should be carried out in the future to further understand the influence of mountain waves or convective activities (e.g., Kim and Chun 2010; Lee and Chun 2018).
Acknowledgments.
This study is funded by the Science and Technology Innovation Program of Hunan Province (2022RC1037) and the National Key Research and Development Program of China (2018YFA0606003). We gratefully acknowledge the European Centre for Medium-Range Weather Forecasts (ECMWF), National Climatic Data Center (NCDC), China Eastern Airlines (CEA), and Ph.D. Zhuge from Nanjing Joint Institute for Atmospheric Sciences (NJIAS) for providing valuable ERA5 reanalysis data, IGRA data, in situ aircraft observation, and AHI cloud-top phase product, respectively. We also would like to thank the reviewers for their insightful and helpful comments.
Data availability statement.
Data are freely available from the corresponding authors by request.
REFERENCES
Chambers, E., 1955: Clear air turbulence and civil jet operations. Aeronaut. J., 59, 613–628, https://doi.org/10.1017/S036839310011692X.
Chang, C., and M. Lu, 2012: Intraseasonal predictability of Siberian high and East Asian winter monsoon and its interdecadal variability. J. Climate, 25, 1773–1778, https://doi.org/10.1175/JCLI-D-11-00500.1.
Doyle, J. D., M. A. Shapiro, Q. Jiang, and D. L. Bartels, 2005: Large-amplitude mountain wave breaking over Greenland. J. Atmos. Sci., 62, 3106–3126, https://doi.org/10.1175/JAS3528.1.
Dudhia, J., 1996: A multi-layer soil temperature model for MM5. Preprints, Sixth PSU/NCAR Mesoscale Model Users Workshop, Boulder, CO, PSU/NCAR, 49–50.
Dutton, J. A., and H. A. Panofsky, 1970: Clear air turbulence: A mystery may be unfolding. Science, 167, 937–944, https://doi.org/10.1126/science.167.3920.937.
Ellrod, G. P., and D. I. Knapp, 1992: An objective clear-air turbulence forecasting technique: Verification and operational use. Wea. Forecasting, 7, 150–165, https://doi.org/10.1175/1520-0434(1992)007<0150:AOCATF>2.0.CO;2.
Endlich, R. M., 1964: The mesoscale structure of some regions of clear-air turbulence. J. Appl. Meteor., 3, 261–276, https://doi.org/10.1175/1520-0450(1964)003<0261:TMSOSR>2.0.CO;2.
Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 1999–2049, https://doi.org/10.1002/qj.3803.
Hu, B., J. Ding, G. Liu, and J. Tang, 2022: Objective verification of clear-air turbulence (CAT) diagnostic performance in China using in-situ aircraft observation. J. Atmos. Oceanic Technol., 39, 903–914, https://doi.org/10.1175/JTECH-D-20-0181.1.
Huang, C., L. Zhou, S. Song, Y. Wu, and S. Liu, 2015: Numerical prediction on clear air turbulence of transoceanic airlines over surrounding waters of South China Sea (in Chinese). J. Meteor. Sci., 35, 317–322.
Iacono, M. J., J. S. Delamere, E. J. Mlawer, M. W. Shephard, S. A. Clough, and W. D. Collins, 2008: Radiative forcing by long-lived greenhouse gases: Calculations with the AER radiative transfer models. J. Geophys. Res., 113, D13103, https://doi.org/10.1029/2008JD009944.
ICAO, 2001: Meteorological Service for International Air Navigation: Annex 3 to the Convention on International Civil Aviation. 14th ed. ICAO, 128 pp.
Janjić, Z. I., 1994: The step-mountain eta coordinate model: Further developments of the convection, viscous sublayer, and turbulence closure schemes. Mon. Wea. Rev., 122, 927–945, https://doi.org/10.1175/1520-0493(1994)122<0927:TSMECM>2.0.CO;2.
Janjić, Z. I., 1996: The surface layer in the NCEP eta model. 11th Conf. on Numerical Weather Predication, Norfolk, VA, Amer. Meteor. Soc., 354–355.
Janjić, Z. I., 2002: Nonsingular implementation of the Mellor–Yamada level 2.5 scheme in the NCEP Meso Model. NCEP Office Note 437, 61 pp., https://repository.library.noaa.gov/view/noaa/11409.
Jhun, J., and E. Lee, 2004: A new East Asian winter monsoon index and associated characteristics of the winter monsoon. J. Climate, 17, 711–726, https://doi.org/10.1175/1520-0442(2004)017<0711:ANEAWM>2.0.CO;2.
Kaplan, M. L., A. W. Huffman, K. M. Lux, J. J. Charney, A. J. Riordan, and Y.-L. Lin, 2005: Characterizing the severe turbulence environments associated with commercial aviation accidents. Part 1: A 44-case study synoptic observational analyses. Meteor. Atmos. Phys., 88, 129–152, https://doi.org/10.1007/s00703-004-0080-0.
Kim, J.-H., and H.-Y. Chun, 2010: A numerical study of clear-air turbulence (CAT) encounters over South Korea on 2 April 2007. J. Appl. Meteor. Climatol., 49, 2381–2403, https://doi.org/10.1175/2010JAMC2449.1.
Kim, J.-H., and H.-Y. Chun, 2012: A numerical simulation of convectively induced turbulence above deep convection. J. Appl. Meteor. Climatol., 51, 1180–1200, https://doi.org/10.1175/JAMC-D-11-0140.1.
Kim, J.-H., H.-Y. Chun, R. D. Sharman, and S. B. Trier, 2014: The role of vertical shear on aviation turbulence within cirrus bands of a simulated western Pacific cyclone. Mon. Wea. Rev., 142, 2794–2813, https://doi.org/10.1175/MWR-D-14-00008.1.
Kim, J.-H., W. N. Chan, B. Sridhar, and R. D. Sharman, 2015: Combined winds and turbulence prediction system for automated air-traffic management applications. J. Appl. Meteor. Climatol., 54, 766–784, https://doi.org/10.1175/JAMC-D-14-0216.1.
Knox, J. A., 1997: Possible mechanism of clear-air turbulence in strongly anticyclonic flow. Mon. Wea. Rev., 125, 1251–1259, https://doi.org/10.1175/1520-0493(1997)125<1251:PMOCAT>2.0.CO;2.
Koch, S. E., and Coauthors, 2005: Turbulence and gravity waves within an upper-level front. J. Atmos. Sci., 62, 3885–3908, https://doi.org/10.1175/JAS3574.1.
Lane, T. P., and R. D. Sharman, 2008: Some influences of background flow conditions on the generation of turbulence due to gravity wave breaking above deep convection. J. Appl. Meteor. Climatol., 47, 2777–2796, https://doi.org/10.1175/2008JAMC1787.1.
Lane, T. P., R. D. Sharman, T. L. Clark, and H.-M. Hsu, 2003: An investigation of turbulence generation mechanisms above deep convection. J. Atmos. Sci., 60, 1297–1321, https://doi.org/10.1175/1520-0469(2003)60<1297:AIOTGM>2.0.CO;2.
Lane, T. P., J. D. Doyle, R. Plougonven, M. A. Shapiro, and R. D. Sharman, 2004: Observations and numerical simulations of inertia-gravity waves and shearing instabilities in the vicinity of a jet stream. J. Atmos. Sci., 61, 2692–2706, https://doi.org/10.1175/JAS3305.1.
Lane, T. P., R. D. Sharman, S. B. Trier, R. G. Fovell, and J. K. Williams, 2012: Recent advances in the understanding of near-cloud turbulence. Bull. Amer. Meteor. Soc., 93, 499–515, https://doi.org/10.1175/BAMS-D-11-00062.1.
Lee, D.-B., and H.-Y. Chun, 2018: A numerical study of aviation turbulence encountered on 13 February 2013 over the Yellow Sea between China and the Korean Peninsula. J. Appl. Meteor. Climatol., 57, 1043–1060, https://doi.org/10.1175/JAMC-D-17-0247.1.
Lee, J. H., J.-H. Kim, R. D. Sharman, J. Kim, and S.-W. Son, 2023: Climatology of clear-air turbulence in upper troposphere and lower stratosphere in the Northern Hemisphere using ERA5 reanalysis data. J. Geophys. Res. Atmos., 128, e2022JD037679, https://doi.org/10.1029/2022JD037679.
Mellor, G. L., and T. Yamada, 1982: Development of a turbulence closure model for geophysical fluid problems. Rev. Geophys., 20, 851–875, https://doi.org/10.1029/RG020i004p00851.
Miles, J. W., 1961: On the stability of heterogeneous shear flows. J. Fluid Mech., 10, 496–508, https://doi.org/10.1017/S0022112061000305.
Miles, J. W., 1986: Richardson’s criterion for the stability of stratified shear flow. Phys. Fluids, 29, 3470–3471, https://doi.org/10.1063/1.865812.
Park, S. H., J.-H. Kim, R. D. Sharman, and J. B. Klemp, 2016: Update of upper level turbulence forecast by reducing unphysical components of topography in the numerical weather prediction model. Geophys. Res. Lett., 43, 7718–7724, https://doi.org/10.1002/2016GL069446.
Roach, W. T., 1970: On the influence of synoptic development on the production of high level turbulence. Quart. J. Roy. Meteor. Soc., 96, 413–429, https://doi.org/10.1002/qj.49709640906.
Sharman, R., and T. Lane, 2016: Aviation Turbulence: Processes, Detection, Prediction. Springer, 523 pp.
Sharman, R., C. Tebaldi, G. Wiener, and J. Wolff, 2006: An integrated approach to mid-and upper-level turbulence forecasting. Wea. Forecasting, 21, 268–287, https://doi.org/10.1175/WAF924.1.
Sharman, R., S. B. Trier, T. P. Lane, and J. D. Doyle, 2012: Sources and dynamics of turbulence in the upper troposphere and lower stratosphere: A review. Geophys. Res. Lett., 39, L12803, https://doi.org/10.1029/2012GL051996.
Skamarock, W. C., and Coauthors, 2008: A description of the Advanced Research WRF version 3. NCAR Tech. Note NCAR/TN-475+STR, 113 pp., https://doi.org/10.5065/D68S4MVH.
Storer, L. N., P. D. Williams, and P. G. Gill, 2019: Aviation turbulence: Dynamics, forecasting, and response to climate change. Pure Appl. Geophys., 176, 2081–2095, https://doi.org/10.1007/s00024-018-1822-0.
Thompson, G., P. R. Field, R. M. Rasmussen, and W. D. Hall, 2008: Explicit forecasts of winter precipitation using an improved bulk microphysics scheme. Part II: Implementation of a new snow parameterization. Mon. Wea. Rev., 136, 5095–5115, https://doi.org/10.1175/2008MWR2387.1.
Trier, S. B., and R. D. Sharman, 2009: Convection-permitting simulations of the environment supporting widespread turbulence within the upper-level outflow of a mesoscale convection system. Mon. Wea. Rev., 137, 1972–1990, https://doi.org/10.1175/2008MWR2770.1.
Trier, S. B., R. D. Sharman, R. G. Fovell, and R. G. Frehlich, 2010: Numerical simulation of radial cloud bands within the upper-level outflow of an observed mesoscale convective system. J. Atmos. Sci., 67, 2990–2999, https://doi.org/10.1175/2010JAS3531.1.
Trier, S. B., R. D. Sharman, and T. P. Lane, 2012: Influences of moist convection on a cold-season outbreak of clear-air turbulence (CAT). Mon. Wea. Rev., 140, 2477–2496, https://doi.org/10.1175/MWR-D-11-00353.1.
Trier, S. B., R. D. Sharman, D. Muñoz-Esparza, and T. L. Keller, 2022: Effects of distant organized convection on forecasts of widespread clear-air turbulence. Mon. Wea. Rev., 150, 2593–2615, https://doi.org/10.1175/MWR-D-22-0077.1.
Williams, P. D., P. L. Read, and T. W. N. Haine, 2003: Spontaneous generation and impact of inertia-gravity waves in a stratified, two layer shear flow. Geophys. Res. Lett., 30, 2255, https://doi.org/10.1029/2003GL018498.
Williams, P. D., T. W. Haine, and P. L. Read, 2005: On the generation mechanisms of short-scale unbalanced modes in rotating two-layer flows with vertical shear. J. Fluid Mech., 528, 1–22, https://doi.org/10.1017/S0022112004002873.
Williams, P. D., T. W. N. Haine, and P. L. Read, 2008: Inertia-gravity waves emitted from balanced flow: Observations, properties, and consequences. J. Atmos. Sci., 65, 3543–3556, https://doi.org/10.1175/2008JAS2480.1.
Zhuge, X., X. Zou, and Y. Wang, 2021: Determining AHI cloud-top phase and intercomparisons with MODIS products over North Pacific. IEEE Trans. Geosci. Remote Sens., 59, 436–448, https://doi.org/10.1109/TGRS.2020.2990955.
