1. Introduction
Near-cloud turbulence (NCT) affecting aviation results from small-scale eddies of 10–1000-m horizontal extent, which occur either outside of clouds or within anvil clouds that are displaced from the regions of active deep convection (e.g., Kim et al. 2014; Lane et al. 2012). Understanding the generation mechanisms for moderate-or-greater (MOG) NCT intensities is important because this turbulence can cause injuries to aircraft crew and passengers (e.g., Kim et al. 2016, 2018; Sharman et al. 2006; Sharman and Pearson 2017) as well as aircraft damage. Commercial aircraft use observations from onboard radar or satellite imagery to help navigate around the areas of deep convection in attempts to avoid turbulence. However, outside of the deep convection aircraft can encounter NCT from a variety of different mechanisms. For instance, at commercial aviation cruising altitudes near the tropopause, the outflow from upstream deep convection can locally intensify the midlatitude jet stream, especially in the northeastern quadrant of mesoscale convective systems (MCSs) in the Northern Hemisphere. This can promote locally strong vertical wind shear that is favorable for NCT (Trier and Sharman 2009; Trier et al. 2020). Above active deep convection regions, breaking of vertically propagating gravity waves can also generate NCT near the critical level or in vertically sheared environments (e.g., Lane et al. 2003; Lane and Sharman 2008; Kim and Chun 2012). In addition, gravity waves that propagate horizontally for several kilometers away from deep convection can perturb background environments with low Richardson number, leading to the turbulence (Fovell et al. 2007; Lane et al. 2012). Due to the lack of complete understanding of all its generation mechanisms, NCT has become a more widely studied topic in recent years in efforts to develop more effective guidelines for turbulence avoidance.
In the current study, several light-to-moderate intensity of turbulence events were observed near the tropical cyclone (TC) Hagibis on 11 October 2019 over the northwestern Pacific Ocean. Hagibis was one of the super typhoons (sustained surface winds of at least 67 m s−1) among the few typhoons (about five) that recurved over East Asia in 2019 and resulted in a widespread disaster with heavy rainfall and extreme winds over Japan (Ito and Ichikawa 2021; Yanase et al. 2022; Jin et al. 2022). Although there were several studies related to the heavy precipitation in this super typhoon, an investigation of the nature of small-scale NCT events that can occur with these systems has not been conducted.
TCs have numerous sources of upper-level turbulence that are potentially hazardous for cruising aircraft over distances of several hundred kilometers or more from the TC center. Cirrus bands, previously identified as a factor inducing NCT in midlatitude MCSs (Trier et al. 2010), extending radially outward from the TC center in its upper-level outflow have been investigated using high-resolution numerical models (Kawashima 2021; Yamazaki and Miura 2021). These studies indicated that the cirrus bands are organized by shallow horizontal convective rolls (HCRs) that form in environments of near-neutral to weak moist static instability and strong vertical wind shear within the TC outflow near the tropopause. Similarly, Kim et al. (2014) argued that vertical wind shear together with the cloud-radiative feedback facilitated NCT of MOG intensity within the radial cirrus bands that were observed in the upper-level outflow of a weakening TC, undergoing transition to an extratropical cyclone. Trier and Sharman (2016) found that convectively induced mesoscale gravity waves were associated with reduced static stability that supported turbulent cirrus banding along the anticyclonic side of a midlatitude jet stream. It was also suggested that the cirrus canopy in the TC is related to the intensification of trapped gravity waves propagating through the upper-level outflow layer of the TC, which is attributed to the diurnal pulse (e.g., Dunion et al. 2014; Navarro et al. 2017; O’Neill et al. 2017; Ditchek et al. 2019).
Other studies have indicated that turbulence can also occur in inertially unstable environments. For instance, Molinari et al. (2019) examined the National Oceanic and Atmospheric Administration (NOAA) Gulfstream-IV (G-IV) aircraft data and found that turbulence can be induced by inertial instability on the northern side of the TC when enhancement of the anticyclonic outflow results from its interaction with environmental westerly winds. Dunkerton (1983) also noted that the small-scale vertical mixings can occur under conditions of weak inertial stability. Scorer (1969) suggested that cellular transverse motion in anticyclonic flow near jet streams could create localized velocity and temperature gradients leading to turbulence. Knox (1997) noted that gravity wave emission resulting from inertial instability in strongly anticyclonic flow can be responsible for generating turbulence. Kim and Chun (2010) showed that strong horizontal gradients of vertical vorticity with local inertial instability on the anticyclonic shear side of the MCS-enhanced midlatitude jet streak can result in horizontal vortex tubes (HVTs; e.g., Clark et al. 2000; Kaplan et al. 2005) with clear-air turbulence. Kim et al. (2014) also reported that observed NCT within radial cirrus bands was collocated with widespread areas of inertial instability due to strong upper-level anticyclonic curvature and shear flows within weakening TC outflows.
The reported turbulence events near Hagibis occurred in the upper-level outflow of the TC, which was located along the southern flank (anticyclonic shear side) of the midlatitude East Asian jet, where inertial stability was weak. In environments of weak inertial stability, vertical mixing can occur in localized regions of inertial instability between the jet streak and the northern side of the TC outflow when they are separated by an optimal horizontal distance to interact with each other (e.g., Rappin et al. 2011; Komaromi and Doyle 2018). Enhancement of the horizontal gradient of potential vorticity when the TC outflow is located equatorward of the jet stream can also cause turbulence in HVTs (Kim and Chun 2010; Doyle et al. 2017; Cowan and Hart 2020). Moreover, strong vertical shear enhanced by the vertical juxtaposition of the intensified environmental jet supported by the TC outflow and background upper-level wind could provide a favorable condition for generating local turbulence.
Several previous studies have identified that the generation mechanisms of observed NCT events related to various instabilities and gravity waves near convective systems were sufficiently resolved and explained in the numerical simulations with horizontal grid spacings less than 1 km (e.g., Lane et al. 2003; Kim and Chun 2010, 2012; Kim et al. 2014; Trier and Sharman 2016; Trier et al. 2020, 2022). Therefore, in this study, we performed high-resolution numerical simulations to investigate whether any of the abovementioned factors related to the TC outflow over East Asia contributed to the observed aviation-scale turbulence events in TC Hagibis. In addition, the structure of anvil cloud with transverse cirrus bands in Hagibis observed in satellite imagery during the study period was examined, which is a prone area for aviation-scale turbulence as noted earlier. Our simulations utilize nested domains having horizontal grid spacings ranging from a few kilometers down to several hundred meters to explicitly represent the downscale cascade processes within synoptic- and mesoscale environments to small-scale eddies in which turbulence occurs (Muñoz-Esparza et al. 2020). Section 2 introduces the description of NCT events observed near the TC. Section 3 explains the design of the numerical experiments. In section 4, an overview of the simulated results is presented, and two different generation mechanisms of simulated turbulence in different locations of the TC outflow are described in section 5. The summary and discussion follow in section 6.
2. Observation of NCT events
From 0840 UTC 11 October 2019 to 0900 UTC 11 October 2019, turbulence events were observed and quantified by in situ eddy dissipation rate (EDR) data from the Aircraft Meteorological Data Relay (AMDAR) database (NOAA/ESRL 2011). They are high-quality data provided by the World Meteorological Organization for scientific research to enhance weather forecasts and support various aviation-related applications (e.g., Frehlich and Sharman 2010; Kim et al. 2020, 2021, 2022; Lee et al. 2022). The area of turbulence was located at z = 11 km above mean sea level (MSL) (≈250 hPa) in the northwestern quadrant of TC Hagibis, approximately 500 km from its center (Figs. 1f and 2a). The in situ reports included light (green dots in Fig. 2) and moderate (red dots in Fig. 2) intensities of turbulence with EDR values of about 0.16 and 0.28 m2/3 s−1, respectively. Here, the criteria for light and moderate intensity ranges are 0.15 ≤ EDR < 0.22 m2/3 s−1 and 0.22 ≤ EDR < 0.34 m2/3 s−1, respectively (Sharman et al. 2014).
Brightness temperature satellite images from Himawari 10.4 μm (band 13) for (a) 1230 UTC 10 Oct 2019, (b) 1630 UTC 10 Oct 2019, and (c) 2030 UTC 10 Oct 2019 and for (d) 0030 UTC 11 Oct 2019, (e) 0430 UTC 11 Oct 2019, and (f) 0840 UTC 11 Oct 2019. Red and green dots in each panel denote moderate and light intensities of observed turbulence at around 0840 UTC in (f), respectively.
Citation: Monthly Weather Review 153, 3; 10.1175/MWR-D-24-0116.1
(a) Sea level pressure (black contours), (b) horizontal wind speed (shading) with wind barbs at 250 hPa, (c) cloud ice water content (shading), and (d) diagnosed EDR calculated by the method in Kim et al. (2018) at 250 hPa on 0800 UTC 11 Oct 2019 obtained from ERA5 data. Red and green dots denote moderate and light intensities of observed turbulence, respectively.
Citation: Monthly Weather Review 153, 3; 10.1175/MWR-D-24-0116.1
Figure 1 shows the infrared satellite images from the Himawari satellite at 10.4-μm wavelength obtained from Chiba University, Japan, from 1230 UTC 10 October 2019 to 0840 UTC 11 October 2019. During this period, Hagibis, which originated in the western Pacific Ocean near Guam, was moving northward. It underwent track recurvature due to the interaction with a midlatitude jet (Ito and Ichikawa 2021), and subsequently weakened with reduction of wind gusts near its center and the size of its eye after 0030 UTC 11 October (Figs. 1d–f). At 0430 UTC, prominent transverse cirrus bands occurred near the northwest edge of the TC outflow (Fig. 1e), which will be discussed in greater detail in subsequent sections of this paper. The collocation of the cirrus banding with the observed turbulence events at 0840 UTC implies that these cloud bands may be closely related to the occurrence of turbulence (Fig. 1f).
The upper-level wind speed and cloud mixing ratio at 250 hPa using the fifth major global reanalysis produced by ECMWF (ERA5) data (Hersbach et al. 2020) are shown in Figs. 2b and 2c, respectively. A southwesterly jet without a distinct upper-level trough was located on the northern side of the TC (Fig. 2b) (Ito and Ichikawa 2021). The turbulence reports occurred where the cloud mixing ratio was small and outside of deep convection located closer to the TC center (Fig. 2c). Therefore, we categorize the reported turbulence events as an NCT case, instead of a more direct or instantaneous impact from the main convection regions in the TC. The large-scale asymmetry in the TC cloud structure, which exhibits a widespread cirrus anvil region north of the TC (Figs. 1 and 2c), indicates that the TC was undergoing extratropical transition (Figs. 1d–f and 2c). According to previous studies, these cloud structures are often accompanied by low inertial stability along the anticyclonic shear side of the jet (Ito and Ichikawa 2021; Yanase et al. 2022), which could also increase the likelihood of turbulence there.
We expect that mechanisms such as shear instability (e.g., Kelvin–Helmholtz or thermal-shear instability) and inertial instability could be responsible for the turbulence that was indicated by the in situ EDR reports during this period. However, it is challenging to determine which factors contributed most to the turbulence encounters using satellite and reanalysis data alone because of limitations in spatial and temporal coverage of satellite data and the inability of coarse reanalysis data to adequately represent the broad range of scales from synoptic to small-scale turbulence. For instance, the ERA5-derived EDR (e.g., Kim et al. 2018) diagnosed MOG turbulence over extensively broad area in the northwestern quadrant of the TC, resulting in numerous false alarms (Fig. 2d). Therefore, to identify the detailed generation mechanisms of localized turbulence events in this TC, we performed a series of high-resolution numerical simulations.
3. Experimental design
Version 4.3 of the WRF-ARW model was used with four one-way nested domains with horizontal grid spacings of 15 (d01), 5 (d02), 1 (d03), and 0.2 (d04) km (Fig. 3). The domains 1 and 2 were centered on the TC, while domains 3 and 4 focused on the region of the cirrus band and observed turbulence encounters, respectively. One-hundred twelve hybrid vertical layers, which had vertical grid spacings decreasing from 280 to 240 m within the z = 8–13-km layer near the altitude of the incident, were applied to all four domains in Fig. 3. The top of the model was 20 hPa, and a 5–km-deep Rayleigh damping layer was situated beneath the model top to mitigate spurious wave reflections.
Configuration of the WRF Model domains with horizontal grid spacings of 15 (d01), 5 (d02), 1 (d03), and 0.2 km (d04). The gray track shows the best track of Hagibis provided by KMA from 0000 UTC 9 Oct to 0000 UTC 13 Oct 2019 with 6-h interval. The simulated track derived from WRF is indicated with blue line from 0000 UTC 11 Oct 2019 to 1200 UTC 11 Oct 2019. Red and green dots denote moderate and light intensities of observed turbulence, respectively.
Citation: Monthly Weather Review 153, 3; 10.1175/MWR-D-24-0116.1
The model was integrated for 12 h from 0000 UTC 11 October 2019 to 1200 UTC 11 October 2019, which encompassed the observed turbulent period. Hourly 0.25° × 0.25° ERA5 data provided initial conditions for each domain and lateral boundary conditions for domain d01. The sea surface temperature was supplemented with Optimum Interpolation Sea Surface Temperature (OISST) data provided by NOAA for every 24 h.
The WSM 6-class microphysics including graupel (Hong and Lim 2006), RRTMG longwave and shortwave radiation (Iacono et al. 2008; Mlawer et al. 1997), and the unified Noah land surface (Ek et al. 2003) parameterization schemes were used in all domains. The Kain–Fritsch cumulus scheme (Kain 2004) was applied only in domain 1. The MYJ planetary boundary layer scheme (Janić 2001) was used in each domain to parameterize local vertical mixings with subgrid-scale turbulent kinetic energy (SGS TKE) in the free atmosphere represented by the Mellor–Yamada 2.5-level turbulence closure method.
Two additional sensitivity simulations were performed to understand the impact of moisture and cloud-radiative feedbacks on storm-induced upper-level outflows closely related to the generation of turbulence. First, a dry (DRY) simulation, removing moist convection and its associated upper-level outflow, was conducted by turning off the microphysics and cumulus parameterization schemes (e.g., Kim et al. 2014; Trier and Sharman 2016; Trier et al. 2020). In DRY, a weak upper-level structure of the TC (e.g., cyclonic circulation near the TC center) may still appear despite the inhibition of TC intensification, since the initial boundary conditions include sea level pressure forcing for the TC. Second, to investigate the effect of longwave and shortwave radiation by ice crystals in anvil cirrus in generating the turbulence, a simulation with no cloud–radiation (NCR) processes was conducted (Trier et al. 2010; Kim et al. 2014). These two sensitivity simulations were otherwise identical to the control (CTL) simulation. Descriptions of the model in this study are summarized in Table 1.
Initial and lateral boundary conditions, resolutions of horizontal domains, vertical layers, and parameterization schemes used in this study.
4. Overview of the simulation results
Figure 4 shows the evolution of model-derived upper-level flow and SGS TKE at 11 km MSL (hereafter all altitudes are MSL), the approximate altitude of the turbulence encounters. The recurving TC was well simulated considering that the cyclonic circulation near the TC center weakened as Hagibis moved northward. The asymmetric structure of the TC outflow along the northwest-to-northern side of the TC also was consistent with the cloud pattern shown in satellite images and ERA5 data (Figs. 1e,f and 2c). The cloud-top temperature values in the upper-level environment, particularly near the observed turbulence region several hundred kilometers away from TC, are also generally consistent with satellite data, which will be shown in Fig. 7 in section 5a. Though the intensity of the TC was underestimated compared to observations (e.g., maximum difference of 20 hPa in the central surface pressure), the simulated track of Hagibis nearly coincided with the best track obtained from the Korean Meteorological Administration (Fig. 3). Simulated synoptic features (e.g., the sea level pressure fields and upper- and lower-tropospheric winds) agree well with the ERA5 analyses, which confirm that both the TC and the background environment were well captured by the simulation. Therefore, we considered the simulation to be reliable for the upper-level outflow-related NCT analysis.
Simulation of horizontal wind speed (shading), wind vectors (black arrows), and SGS TKE (black contour = 0.25 and 0.5 m2 s−2; red contour = 1.0 m2 s−2) for (a) 0330 UTC 11 Oct 2019, (b) 0530 UTC 11 Oct 2019, (c) 0730 UTC 11 Oct 2019, and (d) 0850 UTC 11 Oct 2019 at 11-km altitude in domain 2. The gray shaded region AB indicates the horizontal position of averaged vertical cross sections displayed in Fig. 5. Red and green dots denote moderate and light intensities of observed turbulence at around 0850 UTC in (d), respectively.
Citation: Monthly Weather Review 153, 3; 10.1175/MWR-D-24-0116.1
Before the time of the turbulence incident, widespread areas of SGS TKE larger than 0.25 m2 s−2 were simulated at the altitude of 11 km, especially in the northwestern quadrant of the TC, where the observed turbulence encounters in the anticyclonic outflow from the TC later occurred (Figs. 4a–c). At 0850 UTC (Fig. 4d), SGS TKEs larger than 1.0 m2 s−2 (red contour) occurred at the region of the observed turbulence encounters, implying that local turbulence directly affecting cruising aircraft was also well captured in the simulation.
Horizontally averaged vertical cross sections along AB in Fig. 4 are shown in Fig. 5. At the beginning of the simulation, the low-level structure of the TC in the cyclonic circulation near the eyewall was symmetric, while the upper-level deep convection along the northwestern side of the TC extended farther from the TC center than convection along the southeastern side of the TC (Fig. 5a). The cloud shield of Hagibis, which we will show was associated with inertial instability, progressively became more asymmetric (Figs. 5c,d).
Averaged cross sections of SGS TKE (shading), potential temperature (gray contour with 2-K interval), and cloud mixing ratio (bold gray dashed contour = 0.001 g kg−1) along AB area in Fig. 4 for (a) 0330 UTC 11 Oct 2019, (b) 0530 UTC 11 Oct 2019, (c) 0730 UTC 11 Oct 2019, and (d) 0850 UTC 11 Oct 2019. Red dot denotes moderate intensity of observed turbulence at around 0850 UTC in (d).
Citation: Monthly Weather Review 153, 3; 10.1175/MWR-D-24-0116.1
To identify the evolution of inertial instability during the study period, a vertical cross section of absolute angular momentum (M = u − fy) and cloud mixing ratio is shown in Fig. 6, where u, f, and y represent the wind component perpendicular to section AB, the Coriolis parameter, and the distance from the center of TC, respectively. Decreases (increases) of M in the +y direction denoted in the leftmost panel of Fig. 6 indicates inertially stable (unstable) conditions associated with the TC circulation (e.g., Molinari and Vollaro 2014; Emanuel 1983). At 0330 UTC (Fig. 6a), the eyewall region was inertially stable. In contrast, on the upper and lower flanks of the anvil cloud, where turbulence is later observed (red dot in Fig. 6a), conditions are near neutral to weakly inertially unstable. At later times, the inertial instability becomes more widespread and pronounced within the anvil, as indicated by the overturning of the −40 m s−1 isotach (red contour) and other M surfaces (Figs. 6c,d). This is consistent with previous studies (Rappin et al. 2011; Komaromi and Doyle 2018; Ito and Ichikawa 2021; Yanase et al. 2022), which indicate that asymmetries develop as the inertial stability within the TC weakens under the interaction between the upper-level outflow of TC and midlatitude jet (TC–jet interaction), as in the current case of the East Asian jet located on the north of TC Hagibis (Fig. 2b).
Vertical cross sections of cloud mixing ratio (shading), absolute angular momentum (black contour; red contour for −40 m s−1), and potential temperature (gray contour) along AB line in left figure for (a) 0330 UTC 11 Oct 2019, (b) 0530 UTC 11 Oct 2019, (c) 0730 UTC 11 Oct 2019, and (d) 0850 UTC 11 Oct 2019. The leftmost panel shows the horizontal location of the vertical cross sections along the red solid transect line with red and green dots for moderate and light intensities of observed turbulence at around 0850 UTC. The location of B corresponds to the position of the TC center at each time point.
Citation: Monthly Weather Review 153, 3; 10.1175/MWR-D-24-0116.1
Here, two conditions were needed to allow the TC–jet interaction to drive inertial instability within the TC outflow. The first condition was an optimal horizontal distance between the TC and jet stream, which were close to interact, but far enough to prevent the structure of the TC from disruption by the strong vertical wind shear near the jet, as reported in the idealized simulations of Rappin et al. (2011). The second condition was the presence of an environmental jet located on the north side of the TC core, which provides low inertial stability where these two airstreams can interact. Given the synoptic-scale environment shown in Fig. 2, the separation distance of ∼900 km promoted the TC–jet interaction and the occurrence of inertial instability between the northwestern quadrant of TC and westerly midlatitude jet.
The observed turbulence, which occurred around 0850 UTC, was located near the lower half of the simulated thick TC outflow of 5-km depth indicated by red rectangle in Fig. 5d. Here, it was identified that the turbulence incident at the altitude of 11 km was not directly related to the cirrus band, considering the minimum values of brightness temperature in the satellite image indicate the top of anvil cloud at z = 15 km (Figs. 1f and 5d). Instead, we expect that the strong inertial instability within the overturning of M surfaces at the altitude of the aircraft encounters may have caused the localized NCT, which will be explored further in the next section 5b. In addition to turbulence in the region of the observed NCT events, simulated SGS TKE occurred earlier within the blue rectangle in region 1 (Fig. 5a) and was located at a higher altitude near the upper half of the TC anvil. Thus, the two SGS TKE maxima located at a relatively large distance from the active deep convection are both classified as NCT. However, they arise from different mechanisms. In the next section, we provide a more detailed analysis of the two different mechanisms that result in the onset of NCT in the upper (blue box in Fig. 5a) and lower (red box in Fig. 5d) half of the TC outflow.
5. Two distinct generation mechanisms of NCT within the TC outflow
a. NCT at the top of cirrus outflow canopy
The bands of SGS TKE that occurred near the northwest (NW) edge of the TC outflow at z = 14.5 km in domain 2 with Δx = 5 km (blue box in Fig. 7a), corresponding to the first region of the turbulence at the top of the outflow cloud canopy at 0330 UTC (blue box in Fig. 5a). This area was consistent with the cirrus banding near the observed anvil cloud edge in the satellite imagery (blue box in Fig. 7c). Two sets of resolved cirrus bands revealed by low brightness temperature (Fig. 7b) in the higher-resolution domain 3 (Δx = 1 km) were spatially well correlated with the overall patterns of SGS TKE (blue box in Fig. 7a). HCRs are confirmed as an organizing mechanism (Trier et al. 2010; Kim et al. 2014; Yamazaki and Miura 2021; Kawashima 2021) for the bands, which were substantiated by positive–negative alternating patterns of the x component of horizontal vorticity (∂w/∂y − ∂υ/∂z) oriented parallel to the bands in vertical cross sections (Figs. 7d–f) along the transect AB in Fig. 7b. If cirrus banding arises from Kelvin–Helmholtz (K-H) instability, the phase lines of radial cirrus bands should be perpendicular to the direction of the vertical wind shear vector in the layer. In this case, the phase lines are not perpendicular to the shear vector but are instead parallel to the shear as shown in Fig. 7b, which is similar to that in Kim et al. (2014) and Trier et al. (2010).
Simulated (a) SGS TKE (shading) at 14.5 km in domain 2 and (b) cloud-top brightness temperature (shading) with bulk wind shear vectors (wind barbs) between 15.5 and 12.5 km in domain 3 at 0330 UTC 11 Oct 2019. (c) Brightness temperature satellite imagery from Himawari 10.4 μm (band 13) at 0330 UTC 11 Oct 2019. Vertical cross sections of simulated (d) cloud mixing ratio (shading), (e) SGS TKE (shading), and (f) vertical velocity (shading) with x-component vorticity (red contour; bold and dashed contour for zero and negative values, respectively) and potential temperature (black contour with 2 K interval; bold black contour for 357 K) along AB line in (b). Blue box in (a) and (c) indicates the specific domain displayed in (b).
Citation: Monthly Weather Review 153, 3; 10.1175/MWR-D-24-0116.1
The simulated HCRs developed in a shallow convective layer (i.e., in a region of overturning isentropes; bold black contour in Figs. 7d–f) in the upper half of the anvil cirrus with an approximate 2.5-km depth of the nonzero SGS TKE (Fig. 7e). Locally strong SGS TKE larger than 1.0 m2 s−2 was associated with enhanced vertical motions greater than 2 m s−1 in the HCRs (Fig. 7f), which can result in moderate-to-severe intensity turbulence, capable of directly impacting cruising aircraft as demonstrated in previous studies (e.g., Trier et al. 2010; Kim et al. 2014; Trier and Sharman 2016). The updraft centers of the HCRs (Fig. 7f) led to the formation of the band structure with the maximum cloud ice mixing ratios in the anvil (Fig. 7d), which corresponded to the locations of the minimum brightness temperature along AB in Fig. 7b.
The wind difference magnitude in the layer was larger than 28 m s−1 (≈9 m s−1 km−1) through the 12.5–15.5-km convective layer (Figs. 8a,b). The implied strong vertical shear supports differential thermal advection that leads to static instability and elongated cloud bands associated with HCRs (e.g., Trier et al. 2010; Kim et al. 2014; Yamazaki and Miura 2021). The outermost simulated narrower bulk shear area (Fig. 8a) exhibiting weaker SGS TKE (Fig. 7a), which was collocated with longer cirrus banding with wider spacings between bands at the anvil edge (Fig. 7b), was in the dissipation stage. On the other hand, a broader shear (including green cross mark in Fig. 8a) that included stronger SGS TKE (Fig. 7a) and corresponds to the shorter bands of colder brightness temperature (i.e., banding including transect AB line in Fig. 7b) redeveloped along the inner portion of the NW outflow edge. The redevelopment of the 12.5–15.5-km strong bulk wind shear layer, used as a proxy for the strength of the vertical shear in the upper half of the TC anvil, was found to originate from environmental southwesterlies (Fig. 8a, red barbs) overlying the strong southeasterly radial outflow of the TC (Fig. 8a, blue barbs).
(a) Model-derived maximum reflectivity (shading), wind vectors at 15.5 km (red wind barbs), and wind vectors at 12.5 km (blue wind barbs) and (b) difference of temperature advection between 15.5 and 12.5 km (shading) and wind shear vectors between 15.5 and 12.5 km (black wind barbs) at 0330 UTC 11 Oct 2019 in domain 2. Black contours in (a) and (b) indicate 15.5–12.5-km bulk wind difference magnitude (=28 m s−1). Vertical cross sections of (c) radial wind (shading) and (d) temperature advection (shading) with radial wind vectors (black arrows), potential temperature (gray contours with 2 K interval; white contour for 358 K), and cloud mixing ratio (blue contour = 0.001 g kg−1) along AB line in (a) and (b). Point A in (a) and (b) indicates the center of the TC at 0330 UTC. The horizontal gray dashed lines in (c) and (d) denote the altitudes of 12.5 and 15.5 km. The vertical gray dashed line in (c) and (d) represents the location of green cross in (a) and (b).
Citation: Monthly Weather Review 153, 3; 10.1175/MWR-D-24-0116.1
The core of positive radial wind velocity exceeding 28 m s−1 outward from the TC center was located at 12.5–13.0 km MSL within the cirrus band area (Fig. 8c, gray dashed vertical line), whereas there was near-zero inward flow at 15.5 km in the vertical cross section along AB in Fig. 8a. This strong outflow jet extending from the warm core of deep convection near the TC center induced a maximum in the warm advection in the middle of the cirrus canopy (Fig. 8d). In contrast, weak cold advection from southwesterlies was present above the anvil cloud implying differential thermal advection through the 12.5–15.5-km layer. The area of negative differential temperature advection coincided with the broader band of strong bulk wind shear (Fig. 8b), which supports the idea that onset of cirrus band redevelopment at the inside of the NW outflow edge resulted from differential thermal advection. This differential thermal advection promoted convective instability with isentropic overturning (Figs. 8c,d, white contour), which produced the buoyant forcing and SGS TKE within the top half of the anvil. Finally, the elongated HCRs with cloud bands parallel to the 12.5–15.5-km vertical shear vector were organized in the radial direction and resulted in the eventual stabilization of the convective layer (Figs. 7b and 8b).
In contrast to previous studies (Trier et al. 2010; Kim et al. 2014) investigating cirrus bands aligned beneath the outflow jet of MCSs, the banded cloud features simulated in the current case occurred within the reverse shear located above the TC outflow jet (e.g., Kawashima 2021). Therefore, the axes of the HCRs were oriented antiparallel to the TC outflow jet rather than directed outward from the center of the TC (Figs. 7b and 8b). This result suggests that development conditions for the cirrus bands can vary depending on the intensity and depth of the convective system influencing the strength of the upper-level outflow jet.
Kim et al. (2014) and Yamazaki and Miura (2021) showed cloud-radiative feedbacks to be a crucial factor in either initiating or maintaining cirrus banding in anvils of midlatitude convection and tropical cyclones, respectively. However, other modeling studies have suggested that while cloud-radiative feedbacks contribute, they may not be essential for cirrus banding in some tropical cyclones (e.g., Kawashima 2021) and midlatitude MCSs (e.g., Trier et al. 2010). Neglecting cloud-radiative feedbacks in simulation NCR (Figs. 9b,e) showed relatively minor differences compared to the full physics control simulation (Figs. 9a,d). The redeveloping cirrus band with SGS TKE maxima in the radial direction (Fig. 9e) was reproduced well within the large area of 12.5–15.5-km vertical wind shear in NCR (blue box in Fig. 9b). The largest differences between these two simulations are seen in the outermost banding of low brightness temperature and bulk wind shear implying the role of cloud-radiative feedback in maintaining the strength of the shallow convection within this region.
Model-derived maximum reflectivity (shading), bulk wind difference magnitude (black contour = 28 m s−1) between 15.5 and 12.5 km, wind vectors at 15.5 km (red wind barbs), and wind vectors at 12.5 km (blue wind barbs) for simulations of (a) CTL, (b) NCR, and (c) DRY at 0400 UTC 11 Oct 2019 in domain 2. The cloud-top brightness temperature (shading) with SGS TKE (black contour = 0.1, 0.3, 0.5, and 0.7 m2 s−2) at 15.0 km are shown in simulations (d) CTL, (e) NCR, and (f) DRY at 0400 UTC 11 Oct 2019 in domain 3. Blue box in (a)–(c) indicates the specific domain displayed in (d)–(f).
Citation: Monthly Weather Review 153, 3; 10.1175/MWR-D-24-0116.1
Comparison of DRY (Figs. 9c,f) to CTL (Figs. 9a,b) shows the importance of the strong outflow from the TC at 12.5 km (Fig. 9a, blue barbs) that enhances the vertical shear in the 12.5 to 15.5-km layer (cf. Figs. 9a,c). The significant SGS TKE, which was induced by convective instability arising from differential thermal advection in CTRL and NCR, is not present in DRY due to the absence of strong wind shear (Figs. 9c,f). Thus, strong vertical wind shear due to the upper-level outflow jet from the deep convection had a significant impact on generating the thermal-shear instability at the top of cirrus canopy, while the cloud-radiative feedback had only minor effects on this shallow convective layer of cirrus band in this study.
In this section, the simulated SGS TKE within the cirrus banding at z = 14.5 km in the upper part of the cirrus anvil canopy at the earlier time (0330 UTC) was consistent with the observed banding feature in the satellite data. However, this banding was not collocated with the observed in situ turbulence events at a much lower altitude z = 11 km at a later time (0850 UTC), which could have been generated by a different mechanism and will be addressed in detail in the next section.
b. NCT within the anticyclonic outflow of the TC
At 0850 UTC, SGS TKE values larger than 1.25 m2 s−2 were located within the outflow layer, at the altitude of turbulence encounter (Fig. 10a). The core of the TC outflow jet with speeds over 20 m s−1 at approximately z = 12.0 km was located on the upper-right side of the maximum SGS TKE. The Ri less than 0.25 was widespread in the vicinity of the large SGS TKE layer, which suggests the potential for turbulence arising from shear instability (Fig. 10b), consistent with strong VWS magnitudes exceeding 1.5 × 10−2 s−1 in the lower portion of the simulated anvil cloud (Fig. 10c). A rapid transition of inward radial flow to outward flow below the outflow jet (i.e., through 10–11.5 km) resulted in large vertical wind shear at the lower portion of the anvil cloud as found in previous studies (Trier and Sharman 2009; Molinari et al. 2014). Not surprisingly, moist Brunt–Väisälä frequency (
Vertical cross sections of SGS TKE (shading), potential temperature (gray contours with 2-K interval), radial wind vectors (black arrows), and cloud mixing ratio (blue thin contour = 0.001 g kg−1) with (a) radial wind speed (red dashed contours from 12 to 36 m s−1 at 4 m s−1 interval), (b) Ri (black contour = 0.25), (c) VWS (orange contour = 1.5 × 10−2 s−1), and (d)
Citation: Monthly Weather Review 153, 3; 10.1175/MWR-D-24-0116.1
Simulated AVOR ≤ 0 s−1 (red contour), horizontal wind vectors (black arrows), (a),(c) SGS TKE (shading), and (b)
Citation: Monthly Weather Review 153, 3; 10.1175/MWR-D-24-0116.1
In the vertical cross sections along A–B (Fig. 11a), two branches of southeast-to-northwest-oriented inertial instability were present in portions of the anvil cloud in the z = 8–13-km layer (Fig. 11d). The branch with greater vertical tilt located farther away from the TC center contained convective instability (Fig. 11e) that produced the layers (z = 10–12 km) of SGS TKE stronger than 0.3 m2 s−2 near the observed turbulence event (Fig. 11d). This was the only region where small-scale turbulence occurred under conditions of both inertial (η < 0) and convective (
In the 11–12.5-km layer (31.15°–33.35°N), the radial winds, which were located on the right side of the greatest tilted branch of inertial instability, intensified as they stretched outward and reached their maximum speeds in areas with the weakest inertial stability (Figs. 11d,e). For example, the northwest-oriented outflow core at z = 12.0 km appeared near the layer of the lowest value of η, implying an acceleration of radial outflow into the low inertial stability area (Franklin et al. 1996; Rappin et al. 2011). The coincidence of larger-scale inertial instability and small-scale vertical mixing due to convective instability present only along the outflow channel within the northwestern quadrant of the TC is likely associated with the outward acceleration of the radial wind eventually signifying the presence of instability. The process through which this occurs is discussed in more detail later in this section.
Figure 12 presents simulated time series of
Time series of
Citation: Monthly Weather Review 153, 3; 10.1175/MWR-D-24-0116.1
To identify the detailed evolution of simulated turbulence in the domain 3 (Δx = 1 km), vertical cross sections along AB in Fig. 11b parallel to the radial direction are shown in Fig. 13. At 0712 UTC, there was no significant vertical motion at the observed turbulence encounter locations (Fig. 13d) corresponding to the near-zero SGS TKE region (Fig. 12, vertical line 2). However, small-scale K-H billow features occur near the turbulence area (bold lines in Figs. 13a,d) due to steadily increasing VWS up to 1.13 × 10−2 s−1 (vertical line 2 in Fig. 12). After 0808 UTC (Figs. 13b,e), shallow K-H waves with wavelengths smaller than a few kilometers near the altitude of the observed turbulence excited gravity waves that were trapped in the z = 11–13-km layer and contained vertical motions ranging from −1.3 to 1.3 m s−1, which are similar to structures in other model-based case studies (e.g., Trier et al. 2012, 2022; Trier and Sharman 2016). In the current simulation, there was also a K-H billow-like cloud at z = 13 km within anvil cloud (Fig. 13b), which was collocated with the higher values of VWS that supported nonzero SGS TKE coincident with Ri < 0.25 (vertical line 4 in Fig. 12). Isentropes were overturning in the 11–12-km layer by 0847 UTC (Figs. 13c,f), at which time there was reported turbulence, and large negative values of
Simulated cloud mixing ratio (shading) with equivalent potential temperature (black contours with 1-K interval) for (a) 0712 UTC 11 Oct 2019, (b) 0808 UTC 11 Oct 2019, and (c) 0847 UTC 11 Oct 2019 along AB line in Fig. 11b. Simulated vertical velocity (shading) with equivalent potential temperature (black contours with 1-K interval) for (a) 0712 UTC 11 Oct 2019, (b) 0808 UTC 11 Oct 2019, and (c) 0847 UTC 11 Oct 2019 along AB line in Fig. 11b. Black bold contours at 0712, 0808, and 0847 UTC indicate the 352.6–352.8 (0.2 K interval), 353.2, and 353.4 K equivalent potential temperature, respectively.
Citation: Monthly Weather Review 153, 3; 10.1175/MWR-D-24-0116.1
As shown earlier in Fig. 11, the simulation demonstrates the interrelation between inertial and convective instabilities that was responsible for the reported turbulence events. However, in the time series of Fig. 12, the turbulence region was already in an inertially unstable environment (AVOR < 0) prior to the onset of convective instability (
To further elucidate relationships among these environmental factors in the turbulence event, the vertical profiles of radial wind, tangential wind, SGS TKE, AVOR,
Vertical profiles of (a) radial wind, (b) tangential wind, (c) SGS TKE, (d) AVOR, (e)
Citation: Monthly Weather Review 153, 3; 10.1175/MWR-D-24-0116.1
The increase of radial wind continued during the 0830–0840 UTC time period (Fig. 14a) and was associated with minimum values of the absolute vorticity (Fig. 12). The strong acceleration of radially outward flow (exceeding 12 m s−1) then created convective instability (Fig. 12) and overturning isentropes (Figs. 13c,f). In response to the most enhanced radial flow at 0850 UTC, the strong vertical mixing implied by the SGS TKE maximum (Figs. 12 and 14c) with strengthened convective instability (Figs. 12 and 14e) at a location that coincided with the reported turbulence redistributed vertical momentum resulting in a decrease in the VWS (Figs. 12 and 14f). Vertical mixing also played a role in a reduction of inertial instability after 0840 UTC seen in Fig. 12 (as well as the acceleration of horizontal radial wind), given that the negative absolute vorticity was reduced at the peak of subgrid-scale turbulence (Fig. 12).
Comparison of simulation DRY (Figs. 11c,f) to CTL (Figs. 11a,d) highlights the importance of moist TC radial outflow to generate the simulated NCT near its observed location. Though there was an outflow channel on the southern side of the jet, the inertial instability–induced SGS TKE was not present in DRY due to the absence of acceleration of radial outflow from the TC. For the next 3 h (0900–1200 UTC in Fig. 12), though vertical mixing would act to drive the static and inertial stability, the perturbation patterns of increasing (decreasing) negative absolute vorticity,
A noteworthy aspect of the current case is that vertical mixing occurred within only a thin layer of ∼1-km thickness, as depicted in Figs. 13 and 14. The 11-km MSL altitude of the turbulence reports showed the largest SGS TKE (Fig. 14c), which was simulated in CTL, where
In summary, the reported turbulence occurred within the anticyclonic TC outflow on the northwestern quadrant of Hagibis at low levels of the anvil cloud. Hagibis was in the stage of extratropical transition when it began to be influenced by a background environment of weak inertial stability (yellow zone in Fig. 15) associated with horizontal shear of the East Asian jet (sky blue streamlines in Fig. 15). During this stage, an outflow channel (green streamline in Fig. 15), characterized by strong anticyclonic curvature with localized inertial instability, was formed within the anticyclonic circulation and outflow of the TC (blue streamlines in Fig. 15). This channel extended northeastward toward the low inertial stability region of the environmental jet (yellow zone in Fig. 15).
Schematic feature of environmental flows related to the TC–jet interaction.
Citation: Monthly Weather Review 153, 3; 10.1175/MWR-D-24-0116.1
During the period of turbulence, inertial instability of the outflow channel intensified, which led to stronger horizontal acceleration of the southeasterly radial outflow directed from the TC. Then, the maximized radial outflow in the outflow channel increases the vertical shear triggering KHI, which subsequently evolved into convective instability with overturning of isentropes (red wiggles in Fig. 15). This resulted in the localized strong SGS TKE, which was collocated with encounters of light-or-moderate turbulence at the cruising altitude of aircraft (Fig. 4d). Thus, strong turbulence appears likely to form within the anticyclonic outflow channel in the NW quadrant of TCs when inertial instability–induced KHI or convective instability occurs during the TC–environmental jet interaction. In such instances, the most turbulent region at typical aircraft cruising altitudes may be located several hundred kilometers away from the deep convection core of the TC.
c. Sensitivity of the simulated EDR to the model resolution
Turbulence like that simulated in the current numerical experiments depends strongly on the model resolution due to the small-scale scale nature of the turbulence (Kim and Chun 2012; Kim et al. 2014; Trier and Sharman 2016; Trier et al. 2022). In particular, current research NWP models have horizontal grid spacings that are unable to explicitly resolve the 10–1000-m horizontal scales of turbulent eddies that mostly affect the aircraft. Therefore, SGS TKE, resolved TKE, and EDR have been diagnosed from each model domain (d01–d04 in Fig. 3) to identify the simulated scales of reported turbulence events and assess predictability.
Figure 16a shows the averaged values of SGS TKE, resolved TKE, and total TKE (SGS TKE + resolved TKE) at the locations of light turbulence encounters that were fully contained within the outflow channel (blue dashed boxes in Figs. 17b,c). Here, the resolved TKE
(a) The values of resolved TKE (blue line), SGS TKE (red line), and total TKE (black) line for domains 1, 2, 3, and 4 at 0850 UTC 11 Oct 2019 at 11 km. (b) The values of resolved EDR (blue line), SGS EDR (red line), and total EDR (black) line for domains 1, 2, 3, and 4 at 0850 UTC 11 Oct 2019 at 11 km. The gray dashed line in (b) indicates the observed AMDAR EDR values.
Citation: Monthly Weather Review 153, 3; 10.1175/MWR-D-24-0116.1
Distribution of total EDR (shading) for (a),(b) domain 3 and (c) domain 4 at 11 km at 0850 UTC 11 Oct 2019.
Citation: Monthly Weather Review 153, 3; 10.1175/MWR-D-24-0116.1
To evaluate the accuracy of the simulated fine resolution turbulence events, model-derived EDR was directly compared with the observed in situ EDR data (Fig. 16b). The EDR (=ε1/3) is estimated by ε = q2/3/(b1 × l), where q, b1, and l are the SGS TKE, MYJ coefficient of 11.878 (Muñoz-Esparza et al. 2018), and model-derived mixing length, respectively. The results are similar to those of the TKE (Fig. 16a) with high-resolution domains (d03 and d04) having small values of SGS EDR and large values of resolved EDR (Fig. 16b). The resolved TKE was also the highest in the finest domain (d04). Here, the total EDR (SGS EDR + resolved EDR) in domains 3 and 4 were about 0.25 and 0.3 m2/3 s−1, respectively, which are slightly overestimated but close to the recorded value of in situ AMDAR EDR data (gray dashed line in Fig. 16b). Therefore, in this case, we conclude that the intensity of the observed turbulence encounters affected by large forcing of inertial instability under the TC–environmental jet interaction can be estimated using subkilometer horizontal grid spacing (e.g., d03, d04) where there are significant components of both SGS and resolved EDR.
The simulated spatial patterns of total EDR at 0850 UTC are displayed in Fig. 17. At the aircraft cruising altitude of z = 11 km, total EDRs larger than 0.15 m2/3 s−1 appeared over the northwestern side of TC in domain 3 (Fig. 17a). The simulated light-to-severe turbulence occurred along the outflow channel. In particular, the inset area near the observed turbulence region (Fig. 17b) had locally a high incidence of moderate-to-severe turbulence. In the finest resolution domain 4 (Fig. 17c), severe intensity turbulence occurred at slightly more grid points than in domain 3 (1.08 times as many) and had a fine-scale structure, similar to the results in Trier et al. (2022). These results imply that the turbulence related to inertial instability and KHI can be sufficiently resolved by horizontal grid spacing of 1 and 0.2 km, although the intensity was somewhat higher than the reported data in domain 4 due to much smaller mixing length simulated in the MYJ PBL scheme.
6. Summary and discussion
The generation mechanisms of NCT that occurred within the outflow of TC Hagibis were investigated using high-resolution WRF simulations. Hagibis was a super typhoon that brought devastation to some parts of Japan due to the heavy rainfall and extreme surface winds. Though studies of Hagibis’ precipitation have been conducted, there are no previous studies examining the generation mechanisms of small-scale turbulence produced in the upper-level outflow of the TC. The current study examined the synoptic pattern in which Hagibis was embedded and the downscaling processes lead to local near-tropopause turbulence several hundred kilometers away from the center of the TC.
During the study period, Hagibis moved northward over the northwestern Pacific Ocean and was in a recurvature stage when the eye of the TC became smaller. The anvil cirrus cloud extending to the north and northwest of the TC center had an asymmetric structure, which was collocated with regions of horizontal overturning of absolute momentum surfaces indicating inertial instability associated with the TC–environmental jet interaction.
We examined two distinct areas of simulated NCT at different altitudes, several hundred kilometers from the TC center within the upper-level outflow layer: 1) NCT in the top half of the cirrus anvil cloud and 2) NCT within the anticyclonic outflow of the TC in the bottom half of the anvil cloud, where the first and second areas coincided with observations of cirrus bands in satellite imagery and light and moderate turbulence in situ reports obtained from a commercial airliner, respectively. The agreement between observational data and model outcomes near the turbulence region, located at a distance from the TC core, was considered adequate for analyzing NCT cases using simulated results, despite an underestimation in central pressure for the TC. This discrepancy, however, could not be negligible and may indirectly affect the radial outflow of the TC and small-scale turbulence in local areas, which can be a possible caveat in this study.
In the first location at z = 14 km MSL at earlier times, NCT was simulated in radial cirrus bands that originated from shallow horizontal convective rolls (HCRs). Here, there was strong wind shear between the southerly TC outflow and the overlying environmental southwesterlies, which induced convective instability through differential thermal advection and helped to organize the HCRs. Specifically, the turbulent cirrus bands between 12.5 and 15.5 km MSL were oriented along the vertical shear direction within this layer. However, in contrast to some previous studies, the bands occurred above rather than beneath the maximum along-band winds within the outflow jet. Formation of the cirrus bands in the current case did not depend on the cloud-radiative feedback, which have seemed more important in previous situations (e.g., Kim et al. 2014).
The second location of simulated turbulence that developed later coincided with observed turbulence within the anticyclonic outflow channel at the northwestern quadrant of the TC at 11 km, which was oriented to the south of the jet stream where inertial stability was weak. Before the observed turbulence, the relatively low values of SGS TKE resulted from the strong vertical wind shear, which was conducive to Kelvin–Helmholtz instability (KHI). During the turbulence period, convective instability was triggered and intensified, which resulted in the maximum SGS TKE values. As a result, KHI was attributed to the acceleration of outward radial wind from the TC toward the outflow channel, where the inertial instability is present with negative absolute vorticity. Prior to the period of observed turbulence, intensification of inertial instability within the outflow channel helped accelerate radial flow outward from the TC center toward the direction of the weakest inertial stability. This induced the advection of tangential wind that typically decreases the inertially stable condition. Therefore, the accelerated radial wind, reaching its maximum at the outflow channel above the layer of near-zero radial wind, generated KHI with increasing VWS and subsequently evolved into the convective instability that was responsible for small-scale vertical mixing in an approximate 1-km layer within the half bottom of the anvil cloud. This subgrid-scale turbulence that occurred in the TC–environmental jet interaction region reduced the inertial and static instability, which continued until the end of simulations. These results ultimately demonstrated how inertial instability transitions into local convective instability, supporting earlier arguments of the possibility of turbulence in inertially unstable area within MCSs or TCs (Knox 1997; Kim et al. 2014; Molinari and Vollaro 2014; Molinari et al. 2019). This represents a significant step forward in understanding the turbulence mechanism associated with inertial instability.
NCT that is associated with inertial instability within the anticyclonic outflow of the TC can be partially resolved in simulations with minimum horizontal grid spacings of 1 and 0.2 km. A maximum in total EDRs in this simulation approximately coincided with and had intensity consistent with the reported values of in situ EDR. Because turbulence observations in the upper troposphere and lower stratosphere (UTLS) are limited, comprehensive evaluations of simulated EDR against EDR observations are challenging. More high-resolution simulations of other NCT events are necessary to confirm that convection-allowing numerical weather prediction models with less than 1-km grid spacing can be useful for accurate prediction of the location and intensity of the NCT in the UTLS. This likely depends on the ability of such models to accurately simulate upstream deep convection, which currently varies among different cases. In addition to horizontal resolution, conducting the sensitivity test to higher vertical resolution that can provide more accurate features of turbulence also remains a task to be undertaken in the future, though vertical grid spacing of 250 m in this study sufficiently showed the turbulence with K-H waves and overturning of isentropes within a 1–2-km depth.
Several previous studies have investigated the generation mechanisms of NCT in the outflows of MCSs near the upper-level jet stream using numerical simulation and observational radiosonde data. Moreover, considerable research has been performed on the influences of the interactions between the TC outflow and the environmental jet stream on heavy precipitation. However, there has been far fewer case studies of localized NCT associated with the interaction of the TC outflow with midlatitude jets, especially the East Asian jet. According to the previous climatological studies (Jaeger and Sprenger 2007; Lee et al. 2023; Hu et al. 2023), the northwestern Pacific Ocean region in the vicinity of the East Asian jet is a likely region of clear-air turbulence (CAT) related to inertial instability on the anticyclonic shear side of jet stream in UTLS. The results from the current study should motivate additional case studies of NCT processes and their climatology, which are associated with the interaction between TC outflows and midlatitude jet streams in the future. To support these efforts, we emphasize the need for active field campaigns using research aircraft to obtain the ground-truth records in turbulence-prone areas involving the recurving TCs in the future.
Acknowledgments.
This study was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF), funded by the Ministry of Education (NRF-2019R1I1A2A1060035) and the Korea Meteorological Administration Research and Development Program under Grant KMI2022-00310. James Doyle (JDD) acknowledges the NRL Base Program (PE 61153N) and the Office of Naval Research TC Rapid Intensification DRI (PE 0601153N).
Data availability statement.
The ECMWF reanalysis version 5 hourly data on single and pressure levels used in this study can be accessed at https://cds.climate.copernicus.eu/#!/search?text=&type=dataset. Aircraft Meteorological Data Reports (AMDAR) data also can be obtained from https://data.eol.ucar.edu/dataset/100.016. Data related to simulations are freely available upon request from the corresponding author.
REFERENCES
Barrett, B. S., E. R. Sanabia, S. C. Reynolds, J. K. Stapleton, and A. L. Borrego, 2016: Evolution of the upper tropospheric outflow in Hurricanes Iselle and Julio (2014) in the Navy Global Environmental Model (NAVGEM) analyses and in satellite and dropsonde observations. J. Geophys. Res. Atmos., 121, 13 273–17 286, https://doi.org/10.1002/2016JD025656.
Clark, T. L., W. D. Hall, R. M. Kerr, D. Middleton, L. Radke, F. M. Ralph, P. J. Neiman, and D. Levinson, 2000: Origins of aircraft-damaging clear-air turbulence during the 9 December 1992 Colorado downslope windstorm: Numerical simulations and comparison with observations. J. Atmos. Sci., 57, 1105–1131, https://doi.org/10.1175/1520-0469(2000)057<1105:OOADCA>2.0.CO;2.
Cowan, L. P., and R. E. Hart, 2020: An objective identification and climatology of upper-tropospheric jets near Atlantic tropical cyclones. Mon. Wea. Rev., 148, 3015–3036, https://doi.org/10.1175/MWR-D-19-0262.1.
Ditchek, S. D., J. Molinari, K. L. Corbosiero, and R. G. Fovell, 2019: An objective climatology of tropical cyclone diurnal pulses in the Atlantic basin. Mon. Wea. Rev., 147, 591–605, https://doi.org/10.1175/MWR-D-18-0368.1.
Doyle, J. D., and Coauthors, 2017: A view of tropical cyclones from above: The tropical cyclone intensity experiment. Bull. Amer. Meteor. Soc., 98, 2113–2134, https://doi.org/10.1175/BAMS-D-16-0055.1.
Dunion, J. P., C. D. Thorncroft, and C. S. Velden, 2014: The tropical cyclone diurnal cycle of mature hurricanes. Mon. Wea. Rev., 142, 3900–3919, https://doi.org/10.1175/MWR-D-13-00191.1.
Dunkerton, T. J., 1983: A nonsymmetric equatorial inertial instability. J. Atmos. Sci., 40, 807–813, https://doi.org/10.1175/1520-0469(1983)040<0807:ANEII>2.0.CO;2.
Ek, M. B., K. E. Mitchell, Y. Lin, E. Rogers, P. Grunmann, V. Koren, G. Gayno, and J. D. Tarpley, 2003: Implementation of Noah Land Surface Model advances in the National Centers for Environmental Prediction operational mesoscale Eta model. J. Geophys. Res., 108, 8851, https://doi.org/10.1029/2002JD003296.
Emanuel, K. A., 1983: The Lagrangian parcel dynamics of moist symmetric instability. J. Atmos. Sci., 40, 2368–2376, https://doi.org/10.1175/1520-0469(1983)040<2368:TLPDOM>2.0.CO;2.
Endlich, R. M., 1964: The mesoscale structure of some regions of clear-air turbulence. J. Appl. Meteor. Climatol., 3, 261–276, https://doi.org/10.1175/1520-0450(1964)003<0261:TMSOSR>2.0.CO;2.
Fovell, R. G., R. D. Sharman, and S. B. Trier, 2007: A case study of convectively-induced clear air turbulence. 12th Conf. on Mesoscale Processes, Vol. 13, Waterville Valley, NH, Amer. Meteor. Soc., 13.4, https://ams.confex.com/ams/12meso/techprogram/paper_126190.htm.
Franklin, J. L., S. E. Feuer, J. Kaplan, and S. D. Aberson, 1996: Tropical cyclone motion and surrounding flow relationships: Searching for beta gyres in Omega dropwindsonde datasets. Mon. Wea. Rev., 124, 64–84, https://doi.org/10.1175/1520-0493(1996)124<0064:TCMASF>2.0.CO;2.
Frehlich, R., and R. Sharman, 2010: Climatology of velocity and temperature turbulence statistics determined from rawinsonde and ACARS/AMDAR data. J. Appl. Meteor. Climatol., 49, 1149–1169, https://doi.org/10.1175/2010JAMC2196.1.
Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 1999–2049, https://doi.org/10.1002/qj.3803.
Hong, S.-Y., and J.-O. J. Lim, 2006: The WRF Single-Moment 6-class microphysics scheme (WSM6). Asia-Pac. J. Atmos. Sci., 42, 129–151.
Hu, B., P. Hui, J. Ding, X. Sun, and J. Tang, 2023: Spatiotemporal characteristics of clear-air turbulence (CAT) potential in China during 1979–2020. Climate Dyn., 61, 2339–2353, https://doi.org/10.1007/s00382-023-06684-z.
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.
Ito, K., and H. Ichikawa, 2021: Warm ocean accelerating tropical cyclone Hagibis (2019) through interaction with a mid-latitude westerly jet (in Japanese). SOLA, 17A, 1–6, https://doi.org/10.2151/sola.17A-001.
Jaeger, E. B., and M. Sprenger, 2007: A Northern Hemispheric climatology of indices for clear air turbulence in the tropopause region derived from ERA40 reanalysis data. J. Geophys. Res., 112, D20106, https://doi.org/10.1029/2006JD008189.
Janić, Z. I., 2001: 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.
Jin, H., J. R. Moskaitis, Y. Jin, and J. D. Doyle, 2022: Resolution impact on rapid intensification and structure change of Super Typhoon Hagibis (2019) (in Japanese). J. Meteor. Soc. Japan, 100, 943–964, https://doi.org/10.2151/jmsj.2022-049.
Kain, J. S., 2004: The Kain–Fritsch convective parameterization: An update. J. Appl. Meteor., 43, 170–181, https://doi.org/10.1175/1520-0450(2004)043<0170:TKCPAU>2.0.CO;2.
Kaplan, M. L., A. W. Huffman, K. M. Lux, J. D. Cetola, J. J. Charney, A. J. Riordan, Y.-L. Lin, and K. T. Waight III, 2005: Characterizing the severe turbulence environments associated with commercial aviation accidents. Part 2: Hydrostatic mesoscale numerical simulations of supergradient wind flow and streamwise ageostrophic frontogenesis. Meteor. Atmos. Phys., 88, 153–173, https://doi.org/10.1007/s00703-004-0079-6.
Kawashima, M., 2021: A numerical study of cirrus bands and low-static-stability layers associated with tropical cyclone outflow. J. Atmos. Sci., 78, 3691–3716, https://doi.org/10.1175/JAS-D-21-0047.1.
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, R. D. Sharman, P. D. Williams, and M. Strahan, 2016: Impact of the North Atlantic Oscillation on transatlantic flight routes and clear-air turbulence. J. Appl. Meteor. Climatol., 55, 763–771, https://doi.org/10.1175/JAMC-D-15-0261.1.
Kim, J.-H., R. Sharman, M. Strahan, J. W. Scheck, C. Bartholomew, J. C. H. Cheung, P. Buchanan, and N. Gait, 2018: Improvements in nonconvective aviation turbulence prediction for the World Area Forecast System. Bull. Amer. Meteor. Soc., 99, 2295–2311, https://doi.org/10.1175/BAMS-D-17-0117.1.
Kim, S.-H., H.-Y. Chun, J.-H. Kim, R. D. Sharman, and M. Strahan, 2020: Retrieval of eddy dissipation rate from derived equivalent vertical gust included in Aircraft Meteorological Data Relay (AMDAR). Atmos. Meas. Tech., 13, 1373–1385, https://doi.org/10.5194/amt-13-1373-2020.
Kim, S.-H., H.-Y. Chun, D.-B. Lee, J.-H. Kim, and R. D. Sharman, 2021: Improving numerical weather prediction–based near-cloud aviation turbulence forecasts by diagnosing convective gravity wave breaking. Wea. Forecasting, 36, 1735–1757, https://doi.org/10.1175/WAF-D-20-0213.1.
Kim, S.-H., J. Kim, J.-H. Kim, and H.-Y. Chun, 2022: Characteristics of the derived energy dissipation rate using the 1 Hz commercial aircraft quick access recorder (QAR) data. Atmos. Meas. Tech., 15, 2277–2298, https://doi.org/10.5194/amt-15-2277-2022.
Knox, J. A., 1997: Possible mechanisms of clear-air turbulence in strongly anticyclonic flows. Mon. Wea. Rev., 125, 1251–1259, https://doi.org/10.1175/1520-0493(1997)125<1251:PMOCAT>2.0.CO;2.
Knox, J. A., and V. L. Harvey, 2005: Global climatology of inertial instability and Rossby wave breaking in the stratosphere. J. Geophys. Res., 110, D06108, https://doi.org/10.1029/2004JD005068.
Komaromi, W. A., and J. D. Doyle, 2018: On the dynamics of tropical cyclone and trough interactions. J. Atmos. Sci., 75, 2687–2709, https://doi.org/10.1175/JAS-D-17-0272.1.
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., 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., H.-Y. Chun, S.-H. Kim, R. D. Sharman, and J.-H. Kim, 2022: Development and evaluation of global Korean aviation turbulence forecast systems based on an operational numerical weather prediction model and in situ flight turbulence observation data. Wea. Forecasting, 37, 371–392, https://doi.org/10.1175/WAF-D-21-0095.1.
Lee, J. H., J.-H. Kim, R. D. Sharman, J. Kim, and S.-K. 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.
Mlawer, E. J., S. J. Taubman, P. D. Brown, M. J. Iacono, and S. A. Clough, 1997: Radiative transfer for inhomogeneous atmospheres: RRTM, a validated correlated-k model for the longwave. J. Geophys. Res., 102, 16 663–16 682, https://doi.org/10.1029/97JD00237.
Molinari, J., and D. Vollaro, 2014: Symmetric instability in the outflow layer of a major hurricane. J. Atmos. Sci., 71, 3739–3746, https://doi.org/10.1175/JAS-D-14-0117.1.
Molinari, J., P. Duran, and D. Vollaro, 2014: Low Richardson number in the tropical cyclone outflow layer. J. Atmos. Sci., 71, 3164–3179, https://doi.org/10.1175/JAS-D-14-0005.1.
Molinari, J., M. Rosenmayer, D. Vollaro, and S. D. Ditchek, 2019: Turbulence variations in the upper troposphere in tropical cyclones from NOAA G-IV flight-level vertical acceleration data. J. Appl. Meteor. Climatol., 58, 569–583, https://doi.org/10.1175/JAMC-D-18-0148.1.
Muñoz-Esparza, D., R. D. Sharman, and J. K. Lundquist, 2018: Turbulence dissipation rate in the atmospheric boundary layer: Observations and WRF mesoscale modeling during the XPIA field campaign. Mon. Wea. Rev., 146, 351–371, https://doi.org/10.1175/MWR-D-17-0186.1.
Muñoz-Esparza, D., R. D. Sharman, and S. B. Trier, 2020: On the consequences of PBL scheme diffusion on UTLS wave and turbulence representation in high-resolution NWP models. Mon. Wea. Rev., 148, 4247–4265, https://doi.org/10.1175/MWR-D-20-0102.1.
Navarro, E. L., G. J. Hakim, and H. E. Willoughby, 2017: Balanced response of an axisymmetric tropical cyclone to periodic diurnal heating. J. Atmos. Sci., 74, 3325–3337, https://doi.org/10.1175/JAS-D-16-0279.1.
NOAA/ESRL, 2011: Aircraft Meteorological Data Reports (AMDAR) and Aircraft Communications Addressing and Reporting System (ACARS) Data. Version 1.0. UCAR/NCAR - Earth Observing Laboratory, https://doi.org/10.26023/5FSS-TF0N-V906.
O’Neill, M. E., D. Perez-Betancourt, and A. A. Wing, 2017: Accessible environments for diurnal-period waves in simulated tropical cyclones. J. Atmos. Sci., 74, 2489–2502, https://doi.org/10.1175/JAS-D-16-0294.1.
Rappin, E. D., M. C. Morgan, and G. J. Tripoli, 2011: The impact of outflow environment on tropical cyclone intensification and structure. J. Atmos. Sci., 68, 177–194, https://doi.org/10.1175/2009JAS2970.1.
Scorer, R. S., 1969: Mechanism of clear air turbulence. Clear Air Turbulence and Its Detection, Y.-H. Pao and A. Goldburg, Eds., Plenum Press, 34–50.
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. D., and J. M. Pearson, 2017: Prediction of energy dissipation rates for aviation turbulence. Part I: Forecasting nonconvective turbulence. J. Appl. Meteor. Climatol., 56, 317–337, https://doi.org/10.1175/JAMC-D-16-0205.1.
Sharman, R. D., L. B. Cornman, G. Meymaris, J. Pearson, and T. Farrar, 2014: Description and derived climatologies of automated in situ eddy-dissipation-rate reports of atmospheric turbulence. J. Appl. Meteor. Climatol., 53, 1416–1432, https://doi.org/10.1175/JAMC-D-13-0329.1.
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 convective system. Mon. Wea. Rev., 137, 1972–1990, https://doi.org/10.1175/2008MWR2770.1.
Trier, S. B., and R. D. Sharman, 2016: Mechanisms influencing cirrus banding and aviation turbulence near a convectively enhanced upper-level jet stream. Mon. Wea. Rev., 144, 3003–3027, https://doi.org/10.1175/MWR-D-16-0094.1.
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. P. Lane, 2020: Environment and mechanisms of severe turbulence in a midlatitude cyclone. J. Atmos. Sci., 77, 3869–3889, https://doi.org/10.1175/JAS-D-20-0095.1.
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.
Yamazaki, K., and H. Miura, 2021: On the formation mechanism of cirrus banding: Radiosonde observations, numerical simulations, and stability analyses. J. Atmos. Sci., 78, 3477–3502, https://doi.org/10.1175/JAS-D-20-0356.1.
Yanase, W., K. Araki, A. Wada, U. Shimada, M. Hayashi, and T. Horinouchi, 2022: Multiple dynamics of precipitation concentrated on the North side of Typhoon Hagibis (2019) during extratropical transition (in Japanese). J. Meteor. Soc. Japan, 100, 783–805, https://doi.org/10.2151/jmsj.2022-041.
