VOLUME 49 JOURNAL OF APPLIED METEOROLOGY AND CLIMATOLOGY DECEMBER 2010

A Numerical Study of Clear-Air Turbulence (CAT) Encounters over South

Korea on 2 April 2007

JUNG-HOON KIM AND HYE-YEONG CHUN

Department of Atmospheric Sciences, Yonsei University, Seoul, South Korea

(Manuscript received 20 November 2009, in final form 26 July 2010)

ABSTRACT

Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020

On 2 April 2007, nine cases of moderate-or-greater-level clear-air turbulence (CAT) were observed from
pilot reports over South Korea during the 6.5 h from 0200 to 0830 UTC. Those CAT events occurred in three
different regions of South Korea: the west coast, Jeju Island, and the eastern mountain areas. The charac-
teristics and possible mechanisms of the CAT events in the different regions are investigated using the
Weather Research and Forecasting model. The simulation consists of six nested domains focused on the
Korean Peninsula, with the finest horizontal grid spacing of 0.37 km. The simulated wind and temperature
fields in a 30-km coarse domain are in good agreement with those of the Regional Data Assimilation and
Prediction System (RDAPS) analysis data of the Korean Meteorological Administration and observed
soundings of operational radiosondes over South Korea. In synoptic features, an upper-level front associated
with strong meridional temperature gradients is intensified, and the jet stream passing through the central part
of the Korean Peninsula exceeds 70 m s21. Location and timing of the observed CAT events are reproduced
in the finest domains of the simulated results in three different regions. Generation mechanisms of the CAT
events revealed in the model results are somewhat different in the three regions. In the west coast area, the
tropopause is deeply folded down to about z 5 4 km because of the strengthening of an upper-level front, and
the maximized vertical wind shear below the jet core produces localized turbulence. In the Jeju Island area,
localized mixing and turbulence are generated on the anticyclonic shear side of the enhanced jet, where
inertial instability and ageostrophic flow are intensified in the lee side of the convective system. In the eastern
mountain area, large-amplitude gravity waves induced by complex terrain propagate vertically and sub-
sequently break down over the lee side of topography, causing localized turbulence. For most of the CAT
processes considered, except for the mountain-wave breaking, standard NWP resolutions of tens of kilo-
meters are adequate to capture the CAT events.

1. Introduction CAT forecasting skills and to assist pilots in avoiding

CAT, it is essential to investigate and understand pos-
Clear-air turbulence (CAT) is defined as the bumps of
sible mechanisms of individual CAT events through
in-flight aircraft directly induced by small-scale turbu-
numerical simulations (e.g., Clark et al. 2000; Lane et al.
lent eddies in free atmosphere within a cloud-free or
2003; Lane and Sharman 2008; Trier and Sharman 2009).
stratiform cloud condition (Ellrod et al. 2003). In the
A large portion of CAT has been explained by the
aviation industry, incidents of CAT have been recog-
Kelvin–Helmholtz instability (KHI), which can occur
nized as a serious concern because they occur unex-
when the local Richardson number (Rig 5 N2/VWS2,
pectedly at dominant flight altitudes of commercial
where N and VWS are the Brunt–Väisälä frequency and
aircraft, which result in problems such as occupant in-
vertical wind shear, respectively) is less than 0.25 in
juries, fuel losses, and flight delays. As air transportation
a stably stratified shear flow (Howard 1961; Miles 1961)
has increased, the importance of CAT forecasting has
and less than 1 in a nonlinear 3D flow (Abarbanel et al.
increased (Sharman et al. 2006). To develop more useful
1984, 1986; Miles 1986). Under this condition, turbu-
lent kinetic energy increases by extracting energy from
the mean flow and eventually cascades down to smaller-
Corresponding author address: Prof. Hye-Yeong Chun, Dept.
of Atmospheric Sciences, Yonsei University, 262 Seongsanno, scale eddies, which typically have scales similar to
Seodaemun-gu, Seoul 120-749, South Korea. aircraft (Sekioka 1970; Joseph et al. 2004). In the real
E-mail: chunhy@yonsei.ac.kr atmosphere, CAT events induced by KHI frequently

DOI: 10.1175/2010JAMC2449.1

Ó 2010 American Meteorological Society 2381

2382 JOURNAL OF APPLIED METEOROLOGY AND CLIMATOLOGY
occur above or below the jet stream where the VWS is
sufficiently strong to overcome its environmental stability.
In addition, the entrance region of the jet stream and the
confluence zone of two different flows favor CAT gener-
ation through upper-level frontogenesis (Mancuso and
Endlich 1966; Roach 1970). The strong vertical wind shear
and low static stability near the frontal zone result in KHI.
Ellrod and Knapp (1992) indicated that a turbulence index
1 (TI1 5 DEF 3 VWS, where DEF is total deformation)
that is based on upper-level frontogenesis is a good di-
agnosis for CAT near the developing upper-level front and
the jet core in a highly curved trough. However, it is not
sufficient to describe a highly localized individual CAT
event solely by using the KHI based on frontogenesis (e.g.,
Knox 1997; Kaplan et al. 2005; Ellrod and Knox 2010).
As improvement in observation techniques and com-
puter capacity advances our understanding of CAT mech-
anisms, various other mechanisms of CAT have been
suggested in addition to those related to the jet–front
system. Inertial instability associated with anticyclonic
flow has been considered as an important source of CAT,
although CAT events are less frequent in anticyclonic
flow than in cyclonic flow. Knox (1997) showed that total
deformation induced by upper-level frontogenesis is not
sufficient to explain CAT events in strong anticyclonic
flow. Rather, inertial instability and/or geostrophic ad-
justment could be regarded as a more plausible source
of CAT events in strong anticyclonic flow. Meanwhile,
Clark et al. (2000) demonstrated that the location of the
severe turbulence over Colorado on 9 December 1992 is
consistent with the region of observed and simulated
horizontal vortex tubes (HVTs). In their hypothesis,
these well-mixed and rotating flows of the HVTs are
produced when strong variations in the vertical compo-
nent of relative vorticity transit to the horizontal com-
ponent of vorticity due to flow blocking or convergence
over the terrain. Combining the aforementioned con-
cepts from Knox (1997) and Clark et al. (2000), Kaplan
et al. (2005) suggested a forecasting index, the North
Carolina State University index 2 (NCSU2 5 $z 3 $M,
where z and M are streamwise vertical vorticity and
Montgomery streamfunction on an isentropic surface,
respectively), and showed that the NCSU2 index is max-
imized in a region where the strong gradient of the
streamwise vertical vorticity is perpendicular to the strong
pressure gradient force at an isentropic surface, which fi-
nally transits to the HVTs and causes localized turbulence.
Kaplan et al. (2006) showed that the NCSU2 index can be
a good indicator of severe turbulence in ageostrophi-
cally induced anticyclonic flow regions such as the lee side
of well-developed convective systems. Trier and Sharman
(2009) reported that widespread turbulence indicated by
in situ data from commercial aircraft was associated with
VOLUME 49
strong vertical shear in the upper-level outflow of a
mesoscale convective system (MCS). The subsequent
higher-resolution simulations of Trier et al. (2010) sug-
gested that this observed turbulence may have been re-
lated to shallow banded convection that was organized by
the vertical shear in the near-neutral to weakly con-
vectively unstable outflow.
Another mechanism of CAT is a breakdown of gravity
waves generated by complex terrain and convection. Jiang
and Doyle (2004), Doyle et al. (2005), and Lane et al. (2009)
claimed that large-amplitude mountain waves and their
subsequent breaking over complex terrain are responsible
for small-scale turbulence that directly affects aircraft.
Mountain waves can be a source of CAT in several ways.
First, the amplitude of the vertically propagating mountain
Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020
waves increases with height because of decreasing air
density (Hines 1960; Lindzen 1967), which eventually leads
to wave steepening, overturning, and subsequent breaking
in the higher altitudes (Prusa et al. 1996; Doyle et al. 2005).
Second, an increase in atmospheric stability near the tro-
popause can reduce the vertical wavelength of mountain
waves, which increases the potential for wave breaking
(VanZandt and Fritts 1989). Third, stationary mountain
waves can break down when they approach a critical level
at which the background wind speed equals zero (Lilly
1978; Smith 1979, 1989; Wurtele et al. 1996; Smith et al.
2002). With regard to convective gravity waves, Lane et al.
(2003) suggested that a severe CAT encounter near North
Dakota on 10 July 1997 was related to convectively induced
gravity waves and their subsequent breaking above a well-
developed convective cloud top. Lane and Sharman (2008)
also suggested that CAT induced by convectively generated
gravity waves depends on background stability and VWS
above a well-developed convective cloud.
Climatologically, the region around the Korean Penin-
sula and Japan has a stronger jet stream than any other
region of the world during all seasons, even though the
location of local maxima is different in each season (Koch
et al. 2006). For this reason, potential CAT encounters
diagnosed using several CAT indices are relatively high
over East Asia, including over South Korea, as shown by
Ellrod et al. (2003) and Jaeger and Sprenger (2007). In
addition, the Korean Peninsula is covered mostly (more
than 70%) by complex structures of topography, as shown
in Fig. 1, so that the prevailing westerly flows over South
Korea can produce mountain waves with a wide spectrum.
In these circumstances, the aforementioned generation
mechanisms can be applied to the observed CAT events
over South Korea.
This study documents detailed investigations of
moderate or greater (MOG)-level CAT events over
South Korea on 2 April 2007. In the 6.5 h from 0200 to
0830 UTC, 12 CAT events, including nine MOG-level
DECEMBER 2010 KIM AND CHUN 2383

TABLE 1. Time, latitude, longitude, flight level, and intensity of

nine MOG-level CAT events over South Korea reported in PIREPs
on 2 Apr 2007. Here, W, J, and E in the first column indicate the
three categorized regions: the west coast, Jeju Island, and the
eastern mountain areas of South Korea, respectively. Flight levels
(km) in the fifth column are calculated using the corresponding
standard atmospheric height AGL from the flight altitudes pro-
vided by conventional PIREPs.

Time Lat Lon Flight Turbulence

No. (UTC) (8N) (8E) level (km) intensity
1 (E) 0200 37.08 129.78 11.8 Moderate
2 (J) 0530 33.21 126.48 10.9 Severe
3 (W) 0600 36.80 126.69 7.2 Moderate
4 (W) 0700 37.78 125.70 5.8 Moderate
5 (E) 0730 36.20 129.23 12.2 Moderate
6 (W) 0750 37.58 126.17 5.0 Moderate
7 (E) 0800 36.86 129.46 12.6 Moderate

Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020

8 (J) 0810 32.43 125.73 10.8 Moderate
9 (J) 0830 33.61 126.98 9.8 Moderate

MOG-level CAT events that occurred during the 6.5 h

from 0200 (1100) to 0830 (1730) UTC (LST). Considering
that from 2003 to 2008 the average number of PIREPs and
FIG. 1. Locations of MOG-level CAT events over South Korea on
2 Apr 2007 superimposed on terrain height (shading). The numbers CAT encounters over South Korea has been only 3.4 and
of MOG-level CAT events correspond to those described in Table 1. 2 per day, respectively (Kim and Chun 2011), the occur-
rence of 12 PIREPs and 12 CAT events, with nine MOG-
level CAT events, in a single day is very unusual. The
CAT events, were reported in pilot reports (PIREPs)
atmospheric conditions over the Korean Peninsula during
over three different regions of South Korea: the west
this period might have been particularly vulnerable to gen-
coast, Jeju Island, and the eastern mountain areas. This
erate that much aircraft-scale turbulence. In the present
is the highest single-day frequency of CAT events over
study, we will exclusively investigate the nine MOG-level
South Korea for the last 10 yr (Kim and Chun 2008).
CAT events (CAT events hereafter refers to those nine
The Advanced Research Weather Research and Fore-
MOG-level CAT events). Time, latitude, longitude, flight
casting model (ARW-WRF; Skamarock et al. 2008)
level in km, and turbulence intensity of the CAT events
model with the finest grid spacing of 0.37 km is used to
are presented in Table 1, and their horizontal distribution
reproduce this unique cluster of MOG-level CAT events
is depicted in Fig. 1. As shown in Fig. 1, the CAT events
and to understand mechanisms responsible for them in
are distributed in three regions: the west coast, Jeju Island,
the three different areas of South Korea.
and the eastern mountain areas of South Korea. These
The remainder of this paper is organized as follows. In
three areas are indicated by W, J, and E in Table 1,
section 2, observations of nine MOG-level CAT events
respectively. Table 1 shows that most of the CAT events
over South Korea on 2 April 2007 are described. In
occurred in the 3 h from 0530 to 0830 UTC, except for the
section 3, the experimental setting of the WRF model is
first one that occurred at 0200 UTC. CAT events over the
given. In section 4, synoptic features simulated by the
west coast area occurred within 61 h of 0700 UTC. CAT
WRF model and assimilated in the Regional Data As-
events over Jeju Island and the eastern mountain areas
similation and Prediction System (RDAPS) provided by
have more time gaps between individual CAT events.
the Korea Meteorological Administration (KMA) are
Because commercial aircraft usually measure their alti-
described. In section 5, generation mechanisms for the
tudes by barometers, the altitude of each CAT in Table 1
nine MOG-level CAT events that occurred in the three
is calculated using the standard atmospheric assumption
different regions are investigated in detail. Summary
(Lane et al. 2003; Sharman et al. 2006). Among the re-
and conclusions are given in the last section.
ported altitudes, turbulence levels near the west coast
area are relatively lower (z 5 5–8 km) than in the other
2. Investigation of CAT encounters
two regions (z 5 9–13 km). Among the nine CAT events,
On 2 April 2007, 12 turbulence encounters were re- eight are of moderate intensity, whereas the one at
ported in PIREPs over South Korea. Nine of them were 0530 UTC near Jeju Island is a severe-level CAT.
2384 JOURNAL OF APPLIED METEOROLOGY AND CLIMATOLOGY
FIG. 2. Composite images of MTSAT focused on (a) Korea and (b) East Asia at 0100 UTC 2 Apr 2007.
To identify the existence of clouds during this pe-
riod, composite images of the Multifunctional Transport
Satellite (MTSAT) focused on Korea and East Asia are
used (Fig. 2). In Fig. 2a, there are no clouds over the west
coast and the eastern mountain areas over the Korean
Peninsula, while stratiform clouds that extend from South
China cover Jeju Island. In Fig. 2b, many clouds are lo-
cated along the surface frontal system elongated from
South China to South Japan. Although some patches of
cloud extend to Jeju Island, there are no signals of pre-
cipitation in radar echoes or at the surface observation
stations over South Korea, including Jeju Island on 2 April
2007 (not shown). For these reasons, turbulence that oc-
curred in Korea during this period is regarded as CAT, as
defined by Ellrod et al. (2003).
Before investigating the possible mechanisms of the
CAT events over South Korea, it should be noted that
there are several uncertainties in the PIREP data. First,
the locations of turbulence in PIREPs sometimes dif-
fer by more than hundreds of kilometers from the actual
locations of the turbulence observed by in situ measure-
ments (Cornman et al. 2004). Second, the turbulence
intensity reported in PIREPs tends to be different de-
pending on the particular pilot and aircraft type or size.
The aircraft types of the PIREP data in the present study
are the Boeing 747, Airbus 332, and Airbus 333. Al-
though PIREPs inherently have some uncertainties in the
location and intensity of the turbulence (Schwartz 1996),
they are invaluable resources for validating and evaluat-
ing various CAT diagnostics derived using several oper-
ational forecasting models (e.g., McCann 1999; Lee et al.
2003; Kaplan et al. 2006; Sharman et al. 2006; Knox et al.
2008; Jang et al. 2009; J.-H. Kim et al. 2009).
VOLUME 49
Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020
3. Experimental design of WRF model
The numerical model used in this study is the ARW-
WRF model, version 3.0, developed by the National
Center for Atmospheric Research (NCAR), which is a
nonhydrostatic and fully compressible model. Details of
this model can be found in Skamarock et al. (2008). This
model has been used for several previous studies on
turbulence (e.g., Feltz et al. 2009; Fovell et al. 2007; Trier
and Sharman 2009) and gravity waves (e.g., Plougonven
et al. 2008; S.-Y. Kim et al. 2009). Locations of all model
domains used in the present simulation are defined by
the Lambert conformal map projection shown in Fig. 3.
Six domains (1, 2, 3, 4, 5, and 6) are considered in the
present simulation with horizontal grid spacings of 30,
10, 3.3, 1.1, 1.1, and 0.37 km, respectively. Note that the
horizontal grid spacing of domains 4 and 5 is the same.
The locations of domains 1, 2, and 3 and domains 4, 5,
and 6 are depicted in Figs. 3a and 3b, respectively. Also
shown in Fig. 3b are the locations of the operational
radiosonde stations over South Korea and the nine
MOG-level and three light (LGT)-level CAT events
that occurred on 2 April 2007. Two-way nesting in-
teractions are conducted for domains 1 and 2, 2 and 3, 3
and 4, and 3 and 5, whereas one-way interaction is con-
ducted for domains 5 and 6. Horizontal grid spacing and
size (5730 3 5130 km) of domain 1 are set exactly the
same as those of the operational RDAPS analysis to al-
low direct comparison of synoptic features. Domain 3
includes all CAT events considered in this study, while
domains 4 and 5 encompass those in Jeju Island and the
eastern mountain areas, respectively. Since detailed
examination of the mountain waves and their breaking is
DECEMBER 2010 KIM AND CHUN
FIG. 3. Horizontal locations of (a) the first three domains (domain 1,
Dx 5 30 km; domain 2, Dx 5 10 km; and domain 3, Dx 5 3.3 km)
superimposed on the terrain height (contour interval 5 300 m) of
domain 1 and (b) the last four domains (domain 3; domain 4, Dx 5
1.1 km; domain 5, Dx 5 1.1 km; and domain 6, Dx 5 0.37 km) su-
perimposed on the terrain height (contour interval 5 200 m) of
domain 3. Locations of nine MOG-level and three LGT-level CAT
events are additionally depicted in (b) using the conventional
marks of the given turbulence intensities. Operational radiosonde
stations over South Korea are depicted in (b) as stars (Baekryungdo,
Osan, and Gwangju stations) and dot circles (Sokcho, Pohang,
Heuksando, and Gosan stations).
required, an additional finer-resolution (Dx 5 0.37 km,
where Dx is the horizontal grid spacing) subdomain
(domain 6) is included inside of domain 5.
All domains have 87 vertical layers with a vertical grid
spacing of about 300 m and a model top of 20 hPa (z 5
27–28 km). Rayleigh damping is applied in the uppermost
2385
7 km layer to minimize any artificial reflection. Lateral
boundary layers are specified with 5 relaxation grids in
each domain. For initial and boundary conditions, rean-
alysis data from the National Centers for Environmental
Prediction (NCEP) at 6 h intervals and 18 grid spacing in
latitude and longitude (Derber et al. 1991) are used. The
integrations were conducted for 24 h from 1200 UTC 1 to
1200 UTC 2 April 2007 in domains 1 and 2 and for 18 h
from 1800 UTC 1 to 1200 UTC 2 April 2007 in domains
3–6. For efficiency, model time steps are automatically
determined by an adaptive time-stepping method that
selects time steps for all domains to be as large as possi-
ble to satisfy the Courant–Friedrichs–Lewy (CFL) con-
dition.
The parameterizations used in the present simulation
Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020
include the cloud microphysics scheme by Hong and Lim
(2006), the land surface scheme by Chen and Dudhia
(2001), the Mellor–Yamada–Janjic (MYJ) planetary
boundary layer (PBL) scheme by Janjic (2002), the short
wave radiation scheme by Dudhia (1989), and the long
wave radiation scheme by Mlawer et al. (1997). The cu-
mulus parameterization scheme by Kain (2004) is used
exclusively in domains 1 and 2. Note that the MYJ PBL
scheme takes into account vertical mixing not only in
the PBL but also in the free atmosphere by calculating
subgrid turbulent kinetic energy to third-order accuracy.
Therefore, subgrid-scale turbulence in the CAT regions
can be parameterized by the MYJ PBL scheme without
considering any additional turbulence scheme in the
upper troposphere. To get more realistic model results,
the sixth-order spatial filter (Knievel et al. 2007) is
applied to the model interior, which can damp out 2Dx
waves sufficiently.
4. The synoptic-scale flow
This section describes the synoptic-scale flow pattern
around the Korean Peninsula on 2 April 2007 as it ap-
peared in the RDAPS analysis data and compares it with
the simulated results in the 30-km domain (domain 1).
Figure 4 shows horizontal distributions of wind speed
with geopotential height at the 300-hPa level (Fig. 4a)
and sea level pressure (Fig. 4b) from the RDAPS analysis
(left) and the WRF simulation results (right) at 0000 UTC
2 April 2007. At this time, the jet stream flows from
the west to the east over the Korean Peninsula with
a maximum speed of 74 m s21. In addition, a highly
curved trough in the entrance region of the jet stream
approaches the Korean Peninsula. As expected from
the thermal–wind relationship (›U/›z 5 2›T/›y, where
U and T are the zonal wind and temperature, respec-
tively), the meridional temperature gradient can be
strong over the Korean Peninsula at this time (not
2386 JOURNAL OF APPLIED METEOROLOGY AND CLIMATOLOGY VOLUME 49

Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020

FIG. 4. (a) Geopotential height (gpm; contours) superimposed on horizontal wind speed (m s21; shading) at
300 hPa and (b) sea level pressure (contour) at 0000 UTC 2 Apr 2007 obtained from RDAPS 30-km analysis data
provided by (left) KMA and (right) WRF model output of domain 1. Contour intervals of (a) and (b) are 60 gpm and
4 hPa, respectively.

shown). This is a typical springtime flow pattern in this
area, except for the relatively stronger westerly winds
(about 20 m s21 faster) and highly curved trough at
300 hPa, compared with corresponding fields for the
6-yr (2003–08) springtime mean calculated by the RDAPS
analysis data (not shown). Also, the pressure gradient
force across the Korean Peninsula is exceptionally strong
on 2 April 2007 because of an intensified Siberian high
at the surface. This synoptic feature is similar to that
of CAT events near the upper-level fronts and fronto-
genesis environments shown by previous studies (e.g.,
Roach 1970; Keller 1990; Kaplan et al. 2006). The pre-
vailing wind is almost perpendicular to the meridionally
oriented mountains located in the eastern part of the
Korean Peninsula, and this provides an optimal envi-
ronmental condition to generate large-amplitude moun-
tain waves (Jiang and Doyle 2004; Doyle et al. 2005; Kim
and Chung 2006; Jang and Chun 2008). The simulated
results are fairly good compared with the RDAPS analysis
in both wind and sea level pressure fields. One noticeable
discrepancy, nevertheless, between the analysis and the
simulated is that the simulated center of the Siberian high
shifts southward with somewhat different isobaric lines to
the west of the Siberian high.
For further investigation of the synoptic structure and
for comparison between the model results and obser-
vations, vertical profiles of wind and temperature de-
rived from the simulations in domains 1–3 and observed
from two radiosonde stations are shown in Fig. 5. Be-
cause the simulated soundings in the three domains are
similar, only the result in domain 1 will be discussed and
compared with the observations. Figure 5 shows the
skew T–logp diagrams of temperature, dewpoint tem-
perature, and wind derived using the simulated (gray
DECEMBER 2010 KIM AND CHUN
FIG. 5. Skew T–logp diagrams (solid, temperature; dashed,
dewpoint) obtained from the observed (black) and the simulated
(gray) soundings at (a) Baekryungdo and (b) Gwangju observation
sites in South Korea at 0000 UTC 2 Apr 2007. (right) Simulated and
(far right) observed wind speeds at each level with full (10 m s21)
and half (5 m s21) barbs and a pennant (50 m s21). Isotherms in
108C intervals are denoted by slantwise thin lines in both panels.
lines) and the observed (black lines) soundings at (a)
Baekryungdo and (b) Gwangju at 0000 UTC 2 April
2007. Relatively strong stabilities in the layer between
z 5 4 and 8 km at Baekryungdo and between z 5 3 and
5 km at Gwangju stations are observed. This intensified
stability in the midtroposphere is related to the sub-
sidence of stratospheric air because of deep tropopause
2387
folding. Details of the tropopause folding will be pre-
sented in section 5a. In addition, the maximum wind
speed of 82 (63.5) m s21 is located at 334 (253) hPa
[about z 5 9 (10.4) km] at Baekryungdo (Gwangju)
station, hence strong VWS below the maximum wind
appears between z 5 4 and 8 km at both stations. The
height range of the enhanced VWS is very similar to that
of the CAT events that occurred over the west coast area
of South Korea on 2 April 2007. The relationship be-
tween the enhanced VWS and the CAT events over the
west coast area will also be examined in section 5a. In
general, the model captures well the aforementioned
vertical structures of the observations, although some
discrepancies exist between the simulation and obser-
vations: (i) low-level wind direction and inversion are
Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020
not accurately simulated, and (ii) the simulated max-
imum wind speed at z 5 9 km at Baekryungdo is
72.7 m s21, which is smaller than the observed value
of 82 m s21.
5. Model results
a. West coast area
The potential temperatures in domain 2 at z 5 6.1 km
(average level of number 3, 4, and 6 CAT events) at
1800 UTC 1 (left) and 0700 UTC 2 April 2007 (right) are
shown in Fig. 6a, and horizontal wind speed at z 5 9 km
(the level of the maximum jet stream) with horizontal
wind vectors and pressure at z 5 6.1 km are shown in
Fig. 6b. An intensifying upper-level front and associated
jet stream are identified during this 13-h period. At
1800 UTC, a relatively strong meridional temperature
gradient in a frontal zone is located from Balhae Bay
to the northern part of the Korean Peninsula. As this
frontal zone moves southward, the meridional temper-
ature gradient increases at 0700 UTC (Fig. 6b), implying
an intensified upper-level front during this period. As
the strengthening frontal zone induces strong VWS
through the thermal–wind relationship, the maximum
wind speed of the jet stream above the frontal zone also
increases from 70.2 to 82.9 m s21 over the 13 h. In ad-
dition, the convergence zone of north and south west-
erlies in the entrance region of the jet stream ahead of
the highly curved trough approaches the west coast of
Korea, where the number 3, 4, and 6 CAT events occurred
(Fig. 6b). This flow structure is consistent with several
other CAT events near the upper-level front and the jet
stream (e.g., Keller 1990; Ellrod and Knapp 1992; Kaplan
et al. 2005).
To understand how localized turbulence relates to
the evolution of the intensifying upper-level front/jet
system, we show in Fig. 7 the horizontal wind speed
2388 JOURNAL OF APPLIED METEOROLOGY AND CLIMATOLOGY VOLUME 49

Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020

FIG. 6. Simulation results in domain 2 at (left) 1800 UTC 1 Apr 2007 and (right) 0700 UTC 2 Apr 2007: (a) potential
temperature (contour) at z 5 6.1 km and (b) horizontal wind vectors and pressure (contour) superimposed on
horizontal wind speed (m s21; shading) at z 5 9 km. Contour intervals of (a) and (b) are 4 K and 4 hPa, respectively.
Locations of CAT events over the west coast area are depicted in the right-hand plots as asterisks with numbers
corresponding to those in Table 1.

(Fig. 7a) and Richardson number (Fig. 7b) with poten-
tial temperature and 1.5 potential vorticity unit (PVU)
in y–z cross sections along the line shown in Fig. 6b (right),
using the same data shown in Fig. 6. At 1800 UTC (Fig. 7a,
left), the jet core is located at z 5 9 km and the maxi-
mum wind speed is 74 m s21. The upper-level front that
is represented by slantwise and dense contours of the
potential temperature in the midtroposphere near z 5
3–6 km and y 5 600–1200 km is located below the jet
core and has relatively strong vertical and horizontal
gradients of wind speed and temperature, respectively.
In addition, the height of the dynamic tropopause
(1.5 PVU) decreases from its background height of
about z 5 10–12 km on the anticyclonic side of the jet
core to about z 5 5 km on the cyclonic side following the
upper-level front. At 0700 UTC (Fig. 7b; right), the
overall structure of the upper-level front–jet system is
similar to that at 1800 UTC. However, the slope of the
isentropes near the upper-level front is steeper, so the jet
maximum at z 5 9 km becomes larger (89 m s21). In
addition, the tropopause is deeply folded down to about
z 5 4 km with a steep 1.5 PVU line following the upper-
level front. Because of this deep tropopause folding,
relatively strong stability layers are located in the mid-
troposphere because of the intrusion of relatively stable
stratospheric air. This is consistent with the observation
DECEMBER 2010 KIM AND CHUN 2389

Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020

FIG. 7. Potential temperature (contours) and 1.5 PVU (potential vorticity unit; 1026 K kg21 m2 s21; thick line)
superimposed on (a) horizontal wind speed (m s21; shading) and (b) the Richardson number (shading) along the
dashed line in Fig. 6b (right) at (left) 1800 UTC 1 Apr 2007 and at (right) 0700 UTC 2 Apr 2007. Contour intervals are
4 K. Locations of CAT events over the west coast area are depicted in (b) as asterisks with numbers corresponding to
those in Table 1.

profiles of the temperature shown in Fig. 5. This struc-
ture is also consistent with tropopause folding that oc-
curred along the upper-level frontal zones in previous
studies (e.g., Keyser and Shapiro 1986; Lane et al. 2004;
Koch et al. 2005) and is important not only for aviation
safety but also for the mixing and stratosphere–troposphere
exchange (STE) of chemical constituents (e.g., Shapiro
1980; Pavelin et al. 2001). As the development of the
upper-level front–jet system causes a strong horizon-
tal temperature gradient in the upper-level frontal zone,
VWS is maximized through the thermal wind rela-
tionship. Consequently, the local Richardson number
becomes smaller along this frontal zone, mainly due to
strong VWS regardless of strong stability (see Fig. 7b).
The locations and timing of the number 3, 4, and 6 CAT
events observed over the west coast of Korea are re-
produced in domain 2 of the simulation.
To investigate the resolution dependency of the sim-
ulation results illustrated in Fig. 7b (right), y–z cross
sections of the Richardson number with the potential
temperature and 1.5 PVU in the different domains are
presented in Fig. 8. Figures 8a–8c are the simulated results
from domain 1 (Dx 5 30 km), domain 2 (Dx 5 10 km),
and domain 3 (Dx 5 3.3 km) in the region occupied by
two vertically dashed lines shown in Fig. 7b (right), re-
spectively. Note that the x axis in each plot denotes the
relative distance from the southernmost point of each
domain. Overall features of the tropopause folding and
upper-level frontal system are similar; except that the
lines of 1.5 PVU become steeper as the horizontal grid
resolution becomes higher. In addition, there is a bump
at the dynamic tropopause near x 5 700 km and z 5 5–
6 km of Fig. 8c, caused by relatively large-amplitude
waves generated by steep mountain terrain. The relative
vorticity above the steep mountain near z 5 5–6 km (not
shown) is less than 23 3 1024 s21, and consequently,
potential vorticity has a local minimum there. This is less
evident in low-resolution domains that have smooth and
2390 JOURNAL OF APPLIED METEOROLOGY AND CLIMATOLOGY VOLUME 49

Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020

FIG. 8. Potential temperature (contour), 1.5 PVU (1026 K kg21 m2 s21; thick line), and terrain height (thick line in the bottom) su-
perimposed on the Richardson number (shading) along the region enclosed by two dashed lines at y 5 650 and 1020 km shown in Fig. 7b
(right) in (a) domain 1, (b) domain 2, and (c) domain 3, at 0700 UTC 2 Apr 2007. Contour intervals are 4 K. Note that the x axis in each plot
denotes the relative distance from the southernmost point of each domain. Locations of CAT events over the west coast area are depicted
as asterisks with numbers corresponding to those in Table 1.

small mountains. The regions of the smallest Richardson
number along the upper-level front are located below
the jet core and are mainly due to enhanced VWS along
the intensifying upper-level front/jet system in all three
domains. These regions of small Richardson number
are sustained for about 5 h from 0400 to 0900 UTC,
during which the intensifying upper-level front/jet system
is dominant over the Korean Peninsula. The smallest
Richardson number is 0.224, near the location of the
number 4 CAT in domain 3. In all three domains, the
regions of the small Richardson number are not obvi-
ously different, implying no significant resolution de-
pendency on the simulation results in this area. The
simulation with 30-km horizontal grid spacing can iden-
tify regions of the CAT events sufficiently in this area at
this time.
b. Jeju Island area
In Fig. 7b (right), a region of relatively low static sta-
bility (relatively sparse contours of potential tempera-
ture) and small Richardson number is located on the
anticyclonic shear side of the jet core near z 5 9–12 km
and y 5 300–600 km. The number 2, 8, and 9 CAT events,
observed over the Jeju Island area, are located in this
region at this time in domain 2 (shown in the left of
Fig. 10c, described below). This region has relatively
strong horizontal wind shear but weak vertical wind shear
[vertically oriented contours of isotach; see Fig. 7b (right)].
In some local region on the anticyclonic side of the jet
core, the local Richardson number is less than 0, im-
plying that there exists local vertical mixing of potential
temperature in the left side of Fig. 10c (described below).
Figure 9a shows the accumulated convective precipita-
tion with zero absolute vorticity, horizontal wind vector,
and pressure at the 10.7-km level (averaged level of
the number 2, 8, and 9 CAT events over the Jeju Island
area) derived using the simulated results of domain 1 at
1900 UTC 1 (left) and 0800 UTC 2 (right). Note that Fig. 9a
is not the whole area of domain 1 but focuses on the
regions of anticyclonic flow and the Jeju Island area.
Note also that the absolute vorticity enclosed by the zero
lines is negative. During this period, the Jeju Island area
is located on the anticyclonic shear side of the jet streak
that is passing through the west coast area of South
Korea as described in the previous section. Because of
this strong anticyclonic shear flow, areas of negative
absolute vorticity increase during this period (Fig. 9a).
Considering that the inertial instability associated with
the negative absolute vorticity has been regarded as
a possible source of local turbulence that directly affects
aircraft in the anticyclonic flow (Knox 1997; McCann
2001; Trier and Sharman 2009), CAT events observed in
the Jeju Island area are likely related to the inertial in-
stability. One can apply Stone (1966)’s criterion for
inertial instability—f [ f(1 2 Ri21) 1 z] , 0, where f is
the Coriolis parameter, Ri is the Richardson number,
and z is the vertical component of relative vorticity.
When we apply Stone’s criterion, the areas of inertial
instability are generally similar to those shown in Fig. 9,
DECEMBER 2010 KIM AND CHUN 2391

Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020

FIG. 9. Simulated results in domain 1 at (left) 1900 UTC 1 Apr 2007 and at (right) 0800 UTC 2 April 2007: (a) accumulated convective
precipitation (mm; shading) superimposed with horizontal wind barbs, pressure (contours), and zero absolute vorticity (s21; thick lines) at
z 5 10.7 km and (b) NCSU2 index (310212 s23; shading) superimposed with potential temperature (contour) and 0 absolute vorticity
(thick solid line) along the dashed line in (a). Locations of CAT events over the Jeju Island area are depicted in right panels as asterisks
with numbers corresponding to those in Table 1. Contour intervals of (a) and (b) are 4 hPa and 4 K, respectively.

although they occupy somewhat broader regions. This is
likely because Stone’s criterion includes both symmetric
instability and pure inertial instability.
As the upper-level front–jet system is intensified dur-
ing this period, the maximized wind speed of the jet stream
causes the anticyclonic shear region (negative absolute
vorticity) to be larger near the Jeju Island area (Fig. 9a,
right). On top of that, ageostrophic flow is also domi-
nant, which can be seen from the largely different angle
between the wind vectors and isobaric lines. This strong
ageostrophic flow may be caused by well-developed con-
vective systems in the southern part of China. As seen by
the accumulated convective precipitation in Fig. 9a (right),
a well-developed convective system appears along the
surface frontal zone extending from the mid and southern
parts of China to southern Japan. The overall features of
the simulated precipitation are in good agreement with
the observed convective bands shown in satellite imagery
(Fig. 2b). As dominant westerly flow passes through the
developed convective system, wind vectors are signifi-
cantly perturbed and distorted on the lee side of the
convective system (Fig. 9a, right). As locally enhanced
cross-streamwise confluent flow (ageostrophic flow)
forms the local maximum of the vertical component of
2392 JOURNAL OF APPLIED METEOROLOGY AND CLIMATOLOGY
relative vorticity (vertical vorticity hereafter) on this
anticyclonic shear side of the jet streak region, local
concentration of vertical vorticity causes perturbations
of absolute vorticity. As a result, the zero contours of
absolute vorticity are also perturbed (Fig. 9a, right). As-
suming that strong horizontal gradients of vertical vor-
ticity can be translated to a microscale horizontal vortex
tube (Clark et al. 2000; Kaplan et al. 2005), this could
generate small-scale turbulence that directly affects
aircraft. As can be seen in Fig. 9a (right), the region of
perturbations of zero absolute vorticity along the maxi-
mized streamwise-oriented gradient of vertical vorticity
is consistent with the location of the number 2, 8, and 9
CAT events over the Jeju Island area.
Figure 9b shows the NCSU2 index with zero absolute
vorticity along the dashed line shown in Fig. 9a. The
NCSU2 index is maximized where the vectors between
the pressure gradient force and the gradient of stream-
wise vorticity become orthogonal, in contrast to geo-
strophic flow where those become parallel. At 1900 UTC
1, neither local vertical mixing nor relatively strong val-
ues of the NCSU2 index exist in this cross section, even
though the negative absolute vorticity region is located in
the eastern part of China. As the upper-level front–jet
system intensifies and the dominant westerly flow starts
to be disturbed as it passes through the well-developed
convective system at 0800 UTC 2, local vertical mixing
occurs near z 5 9–11 km and x 5 2500–3300 km. This
region is well matched with the region of locally maxi-
mized values of the NCSU2 index between z 5 6 and
11 km and x 5 2500 and 3300 km, where the number 2, 8,
and 9 CAT events over Jeju Island are located. This
maximized NCSU2 is highly correlated with the negative
absolute vorticity region, which is induced by ageo-
strophic flow on the anticyclonic shear side of the jet
streak, as shown in Fig. 9a. Another indicator to identify
ageostrophic flow is the Lagrangian Rossby number
[Uccellini et al. (1984); RoL 5 jDy/Dtj/jfyj ’ jy  $yj/jfyj,
where y is horizontal wind vector]. The areas of large
values of the RoL (not shown) are found to coexist in the
inertial instability along the lee side of the well-developed
convective clouds, and they are highly consistent with the
areas of large values of NCSU2 index where the number
2, 8, 9 CAT events are located over Jeju Island. The re-
gions of relatively large values of NCSU2 and RoL are
also consistent with the local maximum of subgrid-scale
turbulent kinetic energy (TKE), which will be shown in
Fig. 10a. As a result, both model-diagnosed turbulence
potentials and subgrid-scale TKE capture well the regions
of CAT events over the Jeju Island area.
According to Fig. 9, the well-developed convective
system played an important role in generating the iner-
tial instability along the ageostrophic flow near the number
VOLUME 49
2, 8, and 9 CAT events. To understand the linkage be-
tween the developed convective system and inertial in-
stability, an additional dry simulation is conducted. The
experimental setting of the dry simulation is the same as
for the control simulation, with two-way nesting inte-
gration through domains 1–5, except that convective
parameterization is not used in domains 1 and 2 and
microphysical processes are turned off in all domains.
Figure 10a shows the subgrid-scale TKE with horizontal
wind vectors and boundaries of zero absolute vorticity in
the control simulation with moisture process (left) and
dry simulation (right) at 10.7 km at 0800 UTC 2 April
2007. Note that the light contours in the left figure are
the pressure difference between the control and dry
simulations. Figures 10b and 10c are the same as Figs. 7a
Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020
and 7b, except at 0800 UTC 2 April 2007 in the control
(left) and dry (right) simulations. Several interesting
features can be found in Fig. 10 by comparing the control
and dry simulations. First, the simulated turbulence
(nonzero subgrid-scale TKE) is highly correlated with
the inertial instability in the control simulation, while
there is no turbulence or inertial instability in the dry
simulation near Jeju Island (Fig. 10a). Second, the jet
streak passing through the central part of the Korean
Peninsula is stronger in the control simulation (85 m s21)
than in the dry simulation (60 m s21) at 10.7-km height
(Fig. 10a). This is mainly because the mesohigh on the
top of the MCS intensifies the meridional pressure gra-
dient on its northern side, and Coriolis acceleration
through the geostrophic balance causes the enhanced jet
in the northern or northeastern quadrant of the mesohigh
(Fig. 10a left). Third, comparing the control simulation
with the dry simulation, the jet core occurs at a higher
altitude near y 5 700 km and z 5 10.5 km (Fig. 10b), and
it causes larger horizontal wind shear on the anticyclonic
shear side of the jet core. The minimum Richardson
number associated with inertial instability on the anticy-
clonic shear side of the enhanced jet core occurs in the
upper troposphere where the number 2, 8, and 9 CAT
events are located (Fig. 10c). This is consistent with the
previous study by Trier and Sharman (2009) who dem-
onstrated the relationship between the enhanced jet in
the MCS outflow and aircraft-scale turbulence.
To identify the resolution dependency of the simu-
lated results illustrated in Fig. 10, y–z cross sections of
subgrid-scale TKE superimposed with potential tem-
perature and zero contours of absolute vorticity are
shown in Fig. 11, using the simulated results in domains
1–3 shown in the dashed box in Fig. 10c. Overall features
of the subgrid-scale TKE and potential temperature are
similar in each domain. However, detailed structures
such as the location of the maximum TKE and areas
of negative absolute vorticity are different from each
DECEMBER 2010 KIM AND CHUN 2393

Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020

FIG. 10. Comparison between the (left) control and (right) dry simulations at 0800 UTC 2 Apr, 2007: (a) horizontal wind barbs, pressure
(contours on right), pressure difference between the control and dry (contours on left), and zero absolute vorticity (s21; thick contour)
superimposed on the subgrid-scale TKE (m2 s22; shading) in domain 1, (b) potential temperature (contours) and 1.5 PVU (potential
vorticity unit; 1026 K kg21 m2 s21; thick line) superimposed on horizontal wind speed (m s21; shading), and (c) the Richardson number
(shading) along the dashed line in Fig. 6b (right). Contour intervals in (a) are 1 and 4 hPa, respectively. Contour intervals for (b) and (c)
are 4 K. Locations of CAT events over the Jeju Island area are depicted in (a) and (c) as asterisks with numbers corresponding to those in
Table 1.
2394 JOURNAL OF APPLIED METEOROLOGY AND CLIMATOLOGY VOLUME 49

Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020

FIG. 11. Potential temperature (contours) and zero absolute vorticity (thick line; s21 ) superimposed on the subgrid-scale TKE
(m 2 s 22 ; shading) obtained from the simulated results of (a) domain 1, (b) domain 2, and (c) domain 3 focused on the dashed box
in Fig. 10c. Contour intervals are 4 K. Note that the x axis in each plot is the relative distance from the southernmost point of each
domain. Locations of CAT events over the Jeju Island area are depicted as asterisks with numbers corresponding to those in
Table 1.

other, and small-scale features can be seen as horizontal
grid resolution increases. The maximum subgrid-scale
TKE (1.212 m2 s2) occurs at the location of the number
2 CAT in domain 3, whereas it is less close to the number
2 CAT in domain 1. In all domains, the regions of local
vertical mixing induced by ageostrophically disturbed
flow on the anticyclonic shear side of the jet core are not
obviously different. This suggests that numerical simu-
lation even with 30-km horizontal grid spacing can
identify the regions of CAT events in this area at this
time. In addition, the modeled turbulence in this area is
maintained for about 5 h from 0400 to 0900 UTC in do-
mains 1, 2, 3, and 4 (not shown). This is because the en-
vironmental conditions conducive to CAT continue in
this area as long as the synoptic or mesoscale forcings
support their generation. These results are somewhat
consistent with those of Trier and Sharman (2009), where
the simulated lower Richardson number and the subgrid-
scale TKE with 3-km horizontal and 400-m vertical grid
spacings in the anticyclonic outflow of the MCS are sus-
tained for several hours.
c. Eastern mountain area
This section investigates the generation mechanism
of the CAT events that occurred in the eastern moun-
tain area of South Korea on 2 April 2007. To precisely
simulate mountain waves generated by complex terrain,
topography and large-scale flow must be represented re-
alistically. This is because characteristics of the mountain
waves such as amplitude and wavelength are highly
dependent on the structure of topography as well as up-
stream wind and stability. For this reason, an additional
domain, domain 6 (Dx 5 0.37 km), is included in domain
5 (Dx 5 1.1 km) of the eastern mountain area, as shown in
Fig. 3b, with one-way nesting interaction. Initial condi-
tions and physical parameterizations in domains 5 and 6
are the same as those in the coarser domains, but the
boundary conditions in domain 6 are from the results of
domain 5 updated every 30 min. Large-scale flow during
this period is in good agreement with observations, al-
ready shown in Figs. 4 and 5.
Figure 12 shows vertical velocity (shading) with hor-
izontal wind vectors in domain 3 at four different levels
(z 5 15, 11, 7, and 3 km) at 0300 UTC 2. In this figure,
the localized wave structure is captured over the eastern
side of South Korea. These waves are generated by the
strong zonal flow across the meridionally oriented
mountains over the eastern part of the Korean Penin-
sula. In particular, their amplitudes near the locations of
the number 1, 5, and 7 CAT events over the eastern
mountain area are larger than other regions at z 5
11 km, with the maximum amplitude of vertical velocity
of 13.0 and 24.4 m s21 near the location of the number
7 CAT. This clearly shows that smaller-scale waves ap-
pear more in the lower altitudes, while only components
with relatively long wavelength reach up to the upper
troposphere.
To understand the filtering process of the short-
wavelength waves at low altitude, the vertical profile of
m2 (where m is the vertical wavenumber) for a station-
ary mountain wave is calculated using the following
DECEMBER 2010 KIM AND CHUN 2395

Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020

FIG. 12. Horizontal wind vectors superimposed on vertical velocity (m s21; shading) obtained from the simulated results of domain 3 at
selected heights z 5 (top left) 15, (top right) 11, (bottom left) 7, and (bottom right) 3 km at 0300 UTC 2 Apr 2007. Locations of CAT
events over the eastern mountain area are depicted as asterisks with numbers corresponding to those in Table 1. Location of domain 6 is
indicated as an inner box in the upper-left plot.

equation: m2 5 (N2/U2) 2 (1/U2)(d2U/dz2) 2 k2, where k
is the horizontal wavenumber, and U and N are the basic-
state wind and Brunt–Väisälä frequency, respectively.
Figure 13 shows m2 profiles of waves with horizontal
wavelengths of 12.5 km (Fig. 13a) and 4.2 km (Fig. 13b).
The U and N profiles used to calculate m2 are the domain-
averaged (domain 6) zonal wind and temperature at
0300 UTC 2 April 2007. The vertical wavenumber of the
stationary gravity waves for a given horizontal wave-
length changes with altitude as the background stability
and wind change. Once a vertically propagating gravity
wave meets a level of m2 5 0, it can be reflected backward
from the reflection level (level of m2 5 0) (Gossard and
Hooke 1975). In Fig. 13, a significant reduction in m2
2396 JOURNAL OF APPLIED METEOROLOGY AND CLIMATOLOGY
FIG. 13. Vertical profiles of m2 (where m is the vertical wavenumber of the mountain wave)
with horizontal wavelengths of (a) 12.5 km and (b) 4.2 km calculated using domain-averaged
(domain 6) zonal wind and temperature at 0300 UTC 2 Apr 2007.
appears between z 5 4 and 8 km because of the exis-
tence of strong variance in background stability in the
troposphere, which is mainly due to the deep tropo-
pause folding as mentioned in section 5a. There is no
level of m2 5 0 for the gravity wave with a relatively
longer horizontal wavelength (12.5 km; Fig. 13a), while
a deep evanescent layer exists (z 5 5.2–10.6 km) for the
wave with short horizontal wavelength (4.2 km; Fig.
13b). For the given U and N used in Fig. 13, we found
that waves with horizontal wavelengths shorter than
10.5 km can be reflected in the 4–8-km-height range.
Thus, smaller-scale waves appear more in the lower alti-
tudes, while only relatively long wavelength components
reach up to the upper troposphere, as shown in Fig. 12.
Figure 14 shows x–z cross sections of the Richardson
number (Fig. 14a) and subgrid-scale TKE with potential
temperature (Fig. 14b), zonal wind (Fig. 14c), and ver-
tical velocity (Fig. 14d) in domain 6 at 0300 UTC 2 along
the line of y 5 75 km in domain 6 (36.868N in Fig. 12) at
which the number 7 CAT is located. Zero and negative
values of the wind component are depicted as bold and
dashed lines in Figs. 14c and 14d, respectively. It shows
that mountain waves simultaneously break down in z 5
13.5–17 km. This breaking activates subgrid-scale TKE
in that region (Fig. 14b). However, the simulated tur-
bulence is located above the observed number 7 CAT.
This discrepancy will be examined in Fig. 17, which is
described below. Between z 5 3 and 5 (5 and 9) km,
a relatively strong (weak) stability layer appears, and
this is likely due to the deep tropopause folding, as
shown in Fig. 7. In Fig. 14c, the maximum zonal wind,
which exceeds 70 m s21, locates near z 5 8 km. How-
ever, the zonal wind decreases with height above about z
5 10 km, and 0 wind regions appear near z 5 15 km.
VOLUME 49
Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020
The 0 wind is derived from the breaking of the gravity
waves, and it becomes a wave-induced critical level
(WICL) of the stationary mountain wave (Clark and
Peltier 1984). Once the WICL is formed, consecutive
stationary mountain waves can be either reflected from
or absorbed at the critical level, depending on the local
Richardson number. Since the local Richardson number
in the present case is lower than 0.25 near the WICL
(Fig. 14a), mountain waves are likely to be reflected.
There are two interesting features in Fig. 14d. First,
almost erect or weakly tilted vertical velocities are shown
below about z 5 12 km. Spectral analysis of the vertical
velocity with respect to horizontal wavelength reveals
that there exists a single spectral peak at 22.1 km for all
heights below z 5 12 km (right side of Fig. 16b, described
below). To examine the possibility of ducting between the
surface and z 5 12 km for a wave with a horizontal wave-
length of 22.1 km, the m2 profile of the wave (Fig. 15a) is
calculated using the zonal-mean zonal wind and temper-
ature in the x–z cross section, as shown in Fig. 14d.
Figure 15a shows that although m2 decreases with height
up to about z 5 7 km, it is larger than 0 and increases
above z 5 7 km, with relatively large magnitude at z 5
12 km. We found that a 0-m2 level exists below z 5 12 km
only for waves with horizontal wavelengths less than
7.7 km. This implies that the wave below z 5 12 km with
a 22.1-km horizontal wavelength is not a ducted wave.
Rather, it is a weakly tilted upward-propagating wave.
Figure 15b is the angle of the phase line to the local
vertical for the nonhydrostatic
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi internal gravity wave
given by cosa 5 k/ k 2 1 m2 , which is calculated using
the zonal wavenumber corresponding to the 22.1-km
horizontal wavelength and the vertical wavenumber ob-
tained in Fig. 15a. Figure 15b demonstrates clearly that
DECEMBER 2010 KIM AND CHUN 2397

Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020

FIG. 14. (a) Richardson number (shading) with potential temperature (contours) and subgrid-scale TKE (turbulent kinetic energy;
m2 s22; shading) with (b) potential temperature, (c) zonal wind (contours), and (d) vertical velocity (contours) obtained from the sim-
ulated result of domain 6 at 0300 UTC 2 Apr 2007 along the line of y 5 75 km in domain 6 at which the No. 7 CAT is located. Contour
intervals in (a) and (b) are 4 K and in (c) and (d) are 10 m s21 and 1 m s21, respectively. Zero and negative values of wind velocity in
(c) and (d) are denoted by thick and dashed contours, respectively. Location of No.7 CAT over the eastern mountain area is depicted as
an asterisk with number corresponding to that in Table 1.

the phase angle is almost 0 below z 5 12 km, while it
increases above z 5 12 km as m2 increases. The increase
in m2 above z 5 12 km is mainly due to increases in the
Brunt–Väisälä frequency in the stratosphere. The profiles
of m2 and a calculated using the observed sounding at
Osan (not shown) are similar to those shown in Fig. 15.
This confirms that the waves with horizontal wavelength of
about 22 km below z 5 12 km are vertically propagating
waves rather than ducted waves.
Second, two regions exist of small horizontal wave-
lengths in the layer between z 5 12 and 17 km, one near
x 5 70–100 km and the other near x 5 0–20 km. These
two regions are the wave-breaking regions shown in
Figs. 14a–14c. The waves with horizontal wavelengths
much smaller than those below z 5 12 km are the sec-
ondary waves, and there have been several numerical
modeling studies on this topic associated with mountain
waves (e.g., Bacmeister and Schoeberl 1989; Satomura
and Sato 1999) and convective gravity waves (e.g., Holton
and Alexander 1999; Zhou et al. 2002; Snively and Pasko
2003; Lane and Sharman 2006; Chun and Kim 2008).
Because of the horizontal resolution, the secondary
waves appear only in the two high-resolution sub-
domains (domains 5 and 6) with 1.1- and 0.37-km grid
2398 JOURNAL OF APPLIED METEOROLOGY AND CLIMATOLOGY VOLUME 49

FIG. 15. Vertical profiles of (a) m2 (where m is the vertical wavenumber of the mountain

Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020

wave) and (b) the angle of the wave phase line to the local vertical calculated using the zonal-
mean wind and temperature in the x–z cross section shown in Fig. 14d for a stationary moun-
tain wave with a phorizontal
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi wavelength of 22 km. The angle (a) of the phase line is derived
using cosa 5 k/( k2 1 m2 ), where k and m are the horizontal and vertical wavenumbers,
respectively.

spacings, and waves with horizontal wavelengths less
than about 5 km appear only in domain 6. Shown in
Fig. 16 are the x–z cross sections of vertical velocities
in domains 5 and 6 (left) and their power spectral
densities with respect to horizontal wavelength (right).
The vertical velocity in domain 6 is the one shown in
Fig. 14d. In domain 5 (Fig. 16a), the magnitude of the
mountain waves is very small compared with that in
domain 6, not only for small-scale secondary waves that
are not fully resolvable but also for the dominant wave
with horizontal wavelength of about 22 km below z 5
12 km. The spectral analysis of the vertical velocity in
domain 5 below z 5 12 km reveals that the dominant
horizontal wavelength is 19.4 km (Fig. 16a right), which
is similar to the dominant horizontal wavelength of
22.1 km in domain 6 (Fig. 16b right). Although relatively
small-scale disturbances related to the wave breaking
appear near x 5 140–160 km and z 5 15–17 km in do-
main 5, they are much less evident than in domain 6. As
the horizontal grid spacing decreases, more complex
terrain features becomes resolvable and mountain height
generally increases. Consequently, simulated waves have
a wider spectrum including short wavelength components
and larger amplitudes. Since wave breaking is directly
related to wave amplitude, less or no wave breaking oc-
curs in the relatively low-resolution simulations.
Figure 14 clearly demonstrated that regions of wave
breaking exist in the 13.5–17-km-height range, and this
can generate local turbulence. However, the simulated
wave-breaking regions are located somewhat higher
than the location of the observed number 7 CAT. To
understand the factors that create the discrepancy, we
performed an additional simulation with a different
subgrid-scale turbulence scheme, the Yonsei Univer-
sity (YSU) PBL scheme (Noh et al. 2003; Hong et al.
2006), in domain 6 through one-way nesting to domain
5 that used the MYJ PBL scheme. The simulated tur-
bulence near the number 7 CAT using the YSU PBL in
domain 6, however, is not evidently different from that
using the MYJ PBL scheme (not shown). This implies
that simulated turbulence in domain 6 is not very sensi-
tive to the choice of the subgrid-scale turbulence scheme.
Rather, simulated variables that are important to de-
termine the wave-breaking conditions such as wind and
temperature fields can differ from the observations,
and consequently the simulated wave-breaking height
is higher than the observed height. To examine this
possibility, we estimated the wave-breaking height
from the mountain-drag parameterization by Chun
et al. (1996) based on Palmer et al. (1986), using the
basic-state wind and stability profiles obtained from the
simulation and observations. The following is a brief
description from Chun et al. (1996) of how to estimate
the wave-breaking height: (i) Calculate mountain-wave
stress at the surface (denoted by subscript s) t s 5
2K0rsNsUsjh9j2, where K0 is the horizontal wavenumber
(2.85 3 1024 is used, corresponding to a 22-km horizontal
wavelength), rs is air density, Ns is the Brunt–Väisälä
frequency, Us is the basic-state wind, and jh9j2 is the
variance of subgrid-scale mountain height, which is
calculated in this study using the 30-s digital elevation
model (DEM) dataset at the nearest grid point of the
RDAPS analysis data from the number 7 CAT (h9 5
300 m is used). (ii) Calculate the height variation due
DECEMBER 2010 KIM AND CHUN 2399

Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020

FIG. 16. (left) Vertical velocity (m s21) and (right) vertical profiles of log10-scale power spectrum density (PSD) of
the vertical velocity as a function of zonal horizontal wavelength obtained from the simulated results of (a) domain 5
and (b) domain 6 at 0300 UTC 2 Apr 2007. In the left panels, locations of the No. 7 CAT over the eastern mountain
area are depicted with an asterisk.

to the gravity wave at any level using (dh)k 5 [jtk21j/
(K0rk21Nk21Uk21)]1/2. (iii) Find the wave-breaking
level based on the minimum Richardson number in-
cluding wave effects [Rim 5 Ri(1 2 «)/(1 1 «Ri1/2)2],
where Ri is the local Richardson number, and « is
the inverse Froude number [« 5 (dh)kNk21/Uk21]. If
Rim , 1/ 4, wave breaking occurs and the saturation
wave stress is calculated (Lindzen 1967). If Rim $ 1/ 4, there
is no wave breaking and wave stress is the same as that
at the level below (Eliassen and Palm 1960). (iv) Repeat
the previous step at higher levels until the wave stress is
zero or the model or observation top is reached.
Figure 17 shows vertical profiles of the basic-state wind
(Fig. 17a), Brunt–Väisälä frequency (Fig. 17b), minimum
Richardson number (Fig. 17c), and mountain-wave stress
(Fig. 17d) derived using the simulated result in domain
6 (solid) and an observed rawinsonde sounding at
Osan (dashed) at 0000 UTC 2 April 2007. Osan is lo-
cated upstream of the eastern mountain area and is
depicted in Fig. 3b. The model-produced basic-state
wind and Brunt–Väisälä frequency used in the calcu-
lation of m2 shown in Fig. 15a are obtained from the
zonal-mean zonal wind and temperature in the cross
section shown in Fig. 14. The observed basic-state wind
and Brunt–Väisälä frequency are derived by applying
a nine-point vertical average to the rawinsonde data.
The simulated basic-state wind and stability resembles
the observations in a reasonable sense, except with
relatively weak vertical variation and different mag-
nitude. Because of this discrepancy, vertical profiles of
the minimum Richardson number differ significantly
between the simulated and the observed. Based on the
simulated basic-state flow, the first wave breaking oc-
curs at z 5 14 km, which is consistent with the altitude
of the simulated turbulence shown in Fig. 14. In con-
trast, the wave-breaking height estimated using the
observed basic-state flow in the upper troposphere is
12.8 km, which is close to the altitude of the observed
number 7 CAT (12.6 km). Detailed inspection of the
simulated and observed wind and stability near 12–
14 km reveals that a slight difference in the vertical
variation and magnitude of the basic-state flow causes
a difference in the wave-breaking height of more than
1 km.
2400 JOURNAL OF APPLIED METEOROLOGY AND CLIMATOLOGY VOLUME 49

Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020

FIG. 17. Vertical profiles of (a) U (m s21), (b) N (s21), (c) minimum Richardson number, and
(d) mountain-wave stress (N m22) obtained from the simulated results of domain 6 (solid) and
the observed soundings at Osan station (dashed) at 0000 UTC 2 Apr 2007. Altitudes of wave
breaking estimated from the simulated (solid lines) and the observed (dashed lines) soundings
are depicted as horizontal lines in (c) and (d).

6. Summary and conclusions
On 2 April 2007, nine MOG-level CAT events were
reported in PIREPs during 6.5 h over the Korean Pen-
insula. This is the highest number of CAT occurrences
for this region in one day for the last 10 yr. These CAT
events were distributed in three different regions: the
west coast, Jeju Island, and the eastern mountain areas of
South Korea. Composite images from the MTSAT satel-
lite reveal no clouds over the west coast and eastern
mountain areas, whereas the Jeju Island area is covered
with stratiform clouds extending from the southern part of
China. Based on the available observations such as radar
reflectivity and record of precipitation, all CAT events
including those in Jeju Island occurred in clear or non-
convective weather conditions.
The characteristics and possible generation mechanisms
of the CAT events in the three areas are investigated
using available observations and simulation results of
the WRF model, including six nested domains with the
finest horizontal grid spacing of 0.37 km. Large-scale
features simulated in the lowest-resolution domain (Dx 5
30 km) resemble those in the RDAPS analysis data with
the same resolution. In addition, vertical profiles for wind
direction and speed, temperature, and dewpoint tem-
perature derived from the simulations closely match ra-
diosonde observations.
Based on the simulation results, we found that gen-
eration mechanisms of the CAT events in the three
regions are somewhat different. In the west coast area,
an intensifying upper-level frontal zone is formed by
a strong meridional temperature gradient through the
thermal–wind relationship, and upper-level fronto-
genesis causes strengthening of the jet stream over the
west coast area, which finally introduces maximized
vertical wind shear below the jet core. Deep tropopause
folding on the cyclonic shear side of the jet core leads to the
local Richardson number being smaller than 0.25, which
DECEMBER 2010 KIM AND CHUN
can trigger the Kelvin–Helmholtz instability. The lo-
cations of the KHI are consistent with those of the
observed CAT events over the west coast area. Reso-
lution dependency of the simulated turbulence in this
area is not significant, and a model with 30-km hori-
zontal grid spacing and 300-m vertical grid spacing can
identify the regions of the CAT events during this period.
CAT events observed in the Jeju Island area are likely
associated with inertial instability. As the intensifying
upper-level front/jet system approaches the Korean
Peninsula, the region of inertial instability becomes larger
and moves to the Jeju Island area in the anticyclonic shear
side of the maximized jet core. As strong westerly flows
passing through a well-developed convective system be-
come disturbed, an ageostrophically induced mesoscale
disturbance causes a strong horizontal gradient of local
vertical vorticity in the lee side of the convective system.
This results in the local vertical mixings at z 5 9–11 km
that directly affects aircraft-scale turbulence in the Jeju
Island area. Simulations with 30- and 10-km horizontal
grid spacings with 300-m vertical grid spacing can detect
this turbulence with reliable timing and location.
In the eastern mountain area, CAT events are asso-
ciated with breaking of the mountain waves in the lower
stratosphere. As air density decreases with height, the
amplitudes of the mountain waves become larger above
the jet core, and subsequently mountain waves break
down in a local area and lead to turbulence. Local tur-
bulence due to subsequent wave breaking over the steep-
est mountain area is identified and reproduced in the
simulation result with the finest domain (Dx 5 0.37 km),
even though the altitude of the simulated turbulence is
somewhat higher than the observation. To determine
the reasons for the altitude discrepancy, the height of the
number 7 CAT over the eastern mountain area is esti-
mated using a mountain-wave parameterization with the
observed and simulated soundings. The results show
that a slight difference in the basic-state wind and sta-
bility between the simulation and observations near the
CAT cause a significant discrepancy in the altitudes of
the turbulence (more than 1 km).
Although several discrepancies exist in the location
and timing of individual CAT events between simula-
tion and observations, the generation mechanisms of the
observed CAT events in different regions are properly
reproduced in the present simulations using the WRF
model. It is noteworthy, however, that the capability of
numerical simulation of individual CAT depends not
only on the model resolution but also on its generation
mechanism. CAT events that are generated by large-
scale forcings such as upper-level frontogenesis and in-
ertial instability associated with ageostrophic flow, as
shown in the present west coast and Jeju Island areas,
2401
can be sufficiently identified by current NWP models
with a horizontal grid spacing of ;10–30 km, while those
generated by mountain waves need much higher reso-
lution. This has an important implication for current and
future CAT forecasting. In current CAT forecasting,
based on the combination of various CAT indices cal-
culated using NWP model output, CAT events associ-
ated with mountain waves cannot be diagnosed well, and
thus alternative approaches may be required. For future
CAT forecasting, NWP model–forecasted turbulence
can be used directly instead of calculating diagnostic
CAT indices, given that turbulence generated by large-
scale forcing is explicitly represented in the numeri-
cal simulation with a current NWP model resolution.
As the resolution of NWP models increases, direct
Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020
forecasting of CAT from NWP model is likely to be-
come more beneficial.
To advance our understanding of localized turbulence
in the East Asia region, more CAT cases that occur
under various conditions need to be investigated. In ad-
dition, observational experiments including aircraft in situ
measurements over Korea and East Asia may be needed
to understand the mechanisms of CAT generation, like
the previous studies over the Alps (Jiang and Doyle 2004),
Greenland (Doyle et al. 2005; Ólafsson and Ágústsson
2009), and the Pacific Ocean (Lane et al. 2004; Koch et al.
2005).
Acknowledgments. This work was supported by the
Korea Research Foundation Grant funded by the Ko-
rean Government (MOEHRD, Basic Research Pro-
motion Fund; KRF-2007-313-C00778).
REFERENCES
Abarbanel, H. D. I., D. D. Holm, J. E. Marsden, and T. Ratiu, 1984:
Richardson number criterion for the nonlinear stability of
three-dimensional stratified flow. Phys. Rev. Lett., 52, 2352–2355.
——, ——, ——, and ——, 1986: Nonlinear stability analysis of
stratified fluid equilibria. Philos. Trans. Roy. Soc. London,
A318, 349–409.
Bacmeister, J. T., and M. R. Schoeberl, 1989: Breaking of vertically
propagating two-dimensional gravity waves forced by orog-
raphy. J. Atmos. Sci., 46, 2109–2134.
Chen, F., and J. Dudhia, 2001: Coupling an advanced land surface–
hydrology model with the Penn State–NCAR MM5 modeling
system. Part I: Model implementation and sensitivity. Mon.
Wea. Rev., 129, 569–585.
Chun, H.-Y., and Y.-H. Kim, 2008: Secondary waves generated by
breaking of convective gravity waves in the mesosphere and
their influence in the wave momentum flux. J. Geophys. Res.,
113, D23107, doi:10.1029/2008JD009792.
——, J.-H. Jung, J.-H. Oh, and J.-W. Kim, 1996: Effects of mountain-
induced gravity wave drag on atmospheric general circulation.
J. Korean Meteor. Soc., 32, 581–591.
Clark, T. L., and W. R. Peltier, 1984: Critical level reflection and
the resonant growth of nonlinear mountain waves. J. Atmos.
Sci., 41, 3122–3134.
2402 JOURNAL OF APPLIED METEOROLOGY AND CLIMATOLOGY
——, W. D. Hall, R. M. Kerr, D. Middleton, L. Radke, F. M. Ralph,
P. J. Nieman, and D. Levinson, 2000: Origins of aircraft
damaging clear-air turbulence during the 9 December 1992
Colorado downslope windstorm. J. Atmos. Sci., 57, 1105–1131.
Cornman, L. B., G. Meymaris, and M. Limber, 2004: An update
on the FAA Aviation Weather Research Program’s in situ
turbulence measurement and reporting system. Preprints,
11th Conf. on Aviation, Range, and Aerospace Meteorology,
Hyannis, MA, Amer. Meteor. Soc., P4.3. [Available online at
http://ams.confex.com/ams/11aram22sls/techprogram/paper_
81622.htm.]
Derber, J. C., D. F. Parrish, and S. J. Lord, 1991: The global
operational analysis system at the National Meteorological
Center. Wea. Forecasting, 6, 538–547.
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.
Dudhia, J., 1989: Numerical study of convection observed during
the Winter Monsoon Experiment using a mesoscale two-
dimensional model. J. Atmos. Sci., 46, 3077–3107.
Eliassen, A., and E. Palm, 1960: On the transfer of energy in sta-
tionary mountain waves. Geophys. Publ., 22, 1–23.
Ellrod, G. P., and D. I. Knapp, 1992: An objective clear-air tur-
bulence forecasting technique: Verification and operational
use. Wea. Forecasting, 7, 150–165.
——, and J. A. Knox, 2010: Improvements to an operational clear-
air turbulence diagnostic index by addition of a divergence
trend term. Wea. Forecasting, 25, 789–798.
——, P. F. Lester, and L. J. Ehernberger, 2003: Clear-air turbu-
lence. Encyclopedia of the Atmospheric Sciences, J. R. Holton
et al., Eds., Academic Press, 393–403.
Feltz, W. F., K. M. Bedka, J. A. Otkin, T. Greenward, and
S. A. Ackerman, 2009: Understanding satellite-observed
mountain-wave signatures using high-resolution numerical
model data. Wea. Forecasting, 24, 76–86.
Fovell, R. G., R. D. Sharman, and S. B. Trier, 2007: A case study
of convectively-induced clear-air turbulence. Preprints, 12th
Conf. on Mesoscale Processes, Waterville Valley, NH, Amer.
Meteor. Soc., 13.4. [Available online at http://ams.confex.com/
ams/pdfpapers/126190.pdf.]
Gossard, E. E., and W. H. Hooke, 1975: Waves in the Atmosphere.
Elsevier, 456 pp.
Hines, C. O., 1960: Internal atmospheric gravity waves at iono-
spheric heights. Can. J. Phys., 38, 1441–1481.
Holton, J. R., and M. J. Alexander, 1999: Gravity waves in the
mesosphere generated by tropospheric convection. Tellus,
51B, 45–58.
Hong, S.-Y., and J.-O. J. Lim, 2006: The WRF single-moment 6-class
microphysics scheme (WSM6). J. Korean Meteor. Soc., 42,
129–151.
——, Y. Noh, and J. Dudhia, 2006: A new vertical diffusion
package with an explicit treatment of entrainment processes.
Mon. Wea. Rev., 134, 2318–2341.
Howard, L. N., 1961: Note on a paper by John Miles. J. Fluid Mech.,
10, 509–512.
Jaeger, E. B., and M. Sprenger, 2007: A Northern Hemispheric
climatology of indices for clear-air turbulence in the tropo-
pause region derived from ERA-40 reanalysis data. J. Geo-
phys. Res., 112, D20106, doi:10.1029/2006JD008189.
Jang, W., and H.-Y. Chun, 2008: Severe downslope windstorms
of Gangneung in the springtime. Atmosphere, 18, 207–224.
——, ——, and J.-H. Kim, 2009: A study of forecast system
for clear-air turbulence in Korea. Part I: Korean Integrated
VOLUME 49
Turbulence Forecasting Algorithm (KITFA). Atmosphere,
19, 255–268.
Janjic, Z. I., 2002: Nonsingular implementation of the Mellor-Yamada
level 2.5 scheme in the NCEP Meso Model. NCEP Office Note
437, 61 pp.
Jiang, Q., and J. D. Doyle, 2004: Gravity wave breaking over the
central Alps: Role of complex terrain. J. Atmos. Sci., 61, 2249–
2266.
Joseph, B., A. Mahalov, B. Nicolaenko, and K. L. Tse, 2004: Var-
iability of turbulence and its outer scales in a model tropo-
pause jet. J. Atmos. Sci., 61, 621–643.
Kain, J. S., 2004: The Kain–Fritsch convective parameterization:
An update. J. Appl. Meteor., 43, 170–181.
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 as-
sociated with commercial aviation accidents. Part II: Hydro-
static mesoscale numerical simulations of supergradient wind
Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020
flow and streamwise ageostrophic frontogenesis. Meteor. Atmos.
Phys., 88, 129–153.
——, and Coauthors, 2006: Characterizing the severe turbulence
environments associated with commercial aviation accidents.
A real-time turbulence model (RTTM) designed for the op-
erational prediction of hazardous aviation turbulence envi-
ronments. Meteor. Atmos. Phys., 94, 235–270.
Keller, J. L., 1990: Clear-air turbulence as a response to meso- and
synoptic-scale dynamic processes. Mon. Wea. Rev., 118, 2228–
2242.
Keyser, D., and M. A. Shapiro, 1986: A review of the structure
and dynamics of upper-level frontal zones. Mon. Wea. Rev.,
114, 452–499.
Kim, J.-H., and I.-U. Chung, 2006: Study on mechanisms and
orographic effect for the springtime downslope windstorm
over the Yeongdong region. Atmosphere, 16, 67–83.
——, and H.-Y. Chun, 2008: Analysis on the spatial and temporal
distribution of the aircraft turbulence occurred in South Korea
for the recent 10 years. Preprints, 13th Conf. on Aviation,
Range, and Aerospace Meteorology, New Orleans, LA, Amer.
Meteor. Soc., P3.3. [Available online at http://ams.confex.
com/ams/88Annual/techprogram/paper_128033.htm.]
——, and ——, 2011: Statistics and possible sources of aviation tur-
bulence over South Korea. J. Appl. Meteor. Climatol., in press.
——, ——, W. Jang, and R. D. Sharman, 2009: A study of forecast
system for clear-air turbulence in Korea. Part II: Graphical
Turbulence Guidance (GTG) system. Atmosphere, 19, 269–287.
Kim, S.-Y., H.-Y. Chun, and D. L. Wu, 2009: A study on strato-
spheric gravity waves generated by Typhoon Ewiniar: Nu-
merical simulations and satellite observations. J. Geophys.
Res., 114, D22104, doi:10.1029/2009JD011971.
Knievel, J. C., G. H. Bryan, and J. P. Hacker, 2007: Explicit nu-
merical diffusion in the WRF model. Mon. Wea. Rev., 135,
3808–3824.
Knox, J. A., 1997: Possible mechanism of clear-air turbulence
in strongly anticyclonic flow. Mon. Wea. Rev., 125, 1251–
1259.
——, D. W. McCann, and P. D. Williams, 2008: Application of the
Lighthill–Ford theory of spontaneous imbalance to clear-air
turbulence forecasting. J. Atmos. Sci., 65, 3292–3304.
Koch, P., H. Wernli, and H. W. Davies, 2006: An event-based
jet-stream climatology and typology. Int. J. Climatol., 26,
283–301.
Koch, S. E., and Coauthors, 2005: Turbulence and gravity waves
within an upper-level front. J. Atmos. Sci., 62, 3885–3908.
DECEMBER 2010 KIM AND CHUN
Lane, T. P., and R. D. Sharman, 2006: Gravity wave breaking,
secondary wave generation, and mixing above deep convec-
tion in a three-dimensional cloud model. Geophys. Res. Lett.,
33, L23813, doi:10.1029/2006GL027988.
——, and ——, 2008: Some influences of background flow condi-
tions on the generation of turbulence due to gravity wave
breaking above deep convection. J. Appl. Meteor. Climatol.,
47, 2777–2796.
——, ——, T. L. Clark, and H.-M. Hsu, 2003: An investigation
of turbulence generation mechanisms above deep convection.
J. Atmos. Sci., 60, 1297–1321.
——, 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.
——, ——, R. D. Sharman, M. A. Shapiro, and C. D. Watson, 2009:
Statistics and dynamics of aircraft encounters of turbulence
over Greenland. Mon. Wea. Rev., 137, 2687–2702.
Lee, Y.-G., B.-C. Choi, R. Sharman, G. Wiener, and H.-W. Lee,
2003: Determination of the primary diagnostics for the CAT
(clear-air turbulence) forecast in Korea. J. Korean Meteor.
Soc., 39, 677–688.
Lilly, D. K., 1978: A severe downslope windstorm and aircraft
turbulence event induced by a mountain wave. J. Atmos. Sci.,
35, 59–77.
Lindzen, R. S., 1967: Thermally driven diurnal tide in the atmo-
sphere. Quart. J. Roy. Meteor. Soc., 93, 18–42.
Mancuso, R. L., and R. M. Endlich, 1966: Clear-air turbulence
frequency as a function of wind shear and deformation. Mon.
Wea. Rev., 94, 581–585.
McCann, D. W., 1999: A simple turbulent kinetic energy equa-
tion and aircraft boundary layer turbulence. Natl. Wea. Dig.,
23, 13–19.
——, 2001: Gravity waves, unbalanced flow, and aircraft clear-air
turbulence. Natl. Wea. Dig., 25, 3–14.
Miles, J. W., 1961: On the stability of heteorogeneous shear flows.
J. Fluid Mech., 10, 496–508.
——, 1986: Richardson’s criterion for the stability of stratified
shear flow. Phys. Fluids, 29, 3470–3471.
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.
Noh, Y., W. G. Cheon, S.-Y. Hong, and S. Raasch, 2003: Im-
provement of the K-profile model for the planetary boundary
layer based on large eddy simulation data. Bound.-Layer
Meteor., 107, 401–427.
Ólafsson, H., and H. Ágústsson, 2009: Gravity wave breaking in
easterly flow over Greenland and associated low level barrier-
and reverse tip-jets. Meteor. Atmos. Phys., 104, 191–197.
Palmer, T. N., G. J. Shutts, and R. Swinbank, 1986: Alleviation of a
systematic westerly bias in general circulation and numerical
weather prediction models through an orographic gravity
wave drag parameterization. Quart. J. Roy. Meteor. Soc., 112,
1001–1039.
Pavelin, E., J. A. Whiteway, and G. Vaughan, 2001: Observations
of gravity wave generation and breaking in the lowermost
stratosphere. J. Geophys. Res., 106, 5173–5179.
Plougonven, R., A. Hertzog, and H. Teitelbaum, 2008: Observa-
tions and simulations of a large-amplitude mountain wave
breaking over the Antarctic Peninsula. J. Geophys. Res., 113,
D16113, doi:10.1029/2007JD009739.
2403
Prusa, J. M., P. K. Smolarkiewicz, and R. R. Garcia, 1996: Propa-
gation and breaking at high altitudes of gravity waves excited
by tropospheric forcing. J. Atmos. Sci., 53, 2186–2216.
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.
Satomura, T., and K. Sato, 1999: Secondary generation of gravity
waves associated with the breaking of mountain waves. J. Atmos.
Sci., 56, 3847–3858.
Schwartz, B., 1996: The quantitative use of PIREPs in developing avi-
ation weather guidance products. Wea. Forecasting, 11, 372–384.
Sekioka, M., 1970: Application of Kelvin–Helmholtz instability to
clear-air turbulence. J. Appl. Meteor., 9, 896–899.
Shapiro, M. A., 1980: Turbulent mixing within tropopause folds
as a mechanism for the exchange of chemical constituents
between the stratosphere and troposphere. J. Atmos. Sci., 37,
994–1004.
Downloaded from http://journals.ametsoc.org/doi/pdf/10.1175/2010JAMC2449.1 by guest on 17 June 2020
Sharman, R., C. Tebaldi, G. Wiener, and J. Wolff, 2006: An in-
tegrated approach to mid- and upper-level turbulence fore-
casting. Wea. Forecasting, 21, 268–287.
Skamarock, W. C., J. B. Klemp, J. Dudhia, D. O. Gill, D. M. Barker,
M. G. Duda, X.-Y. Huang, W. Wang, and J. G. Powers, 2008: A
description of the Advanced Research WRF version 3. NCAR
Tech. Note NCAR/TN-4751STR, 88 pp.
Smith, R. B., 1979: The influence of mountains on the atmosphere.
Advances in Geophysics, Vol. 21, Academic Press, 87–230.
——, 1989: Hydrostatic air-flow over mountains. Advances in
Geophysics, Vol. 31, Academic Press, 1–41.
——, S. Skubis, J. D. Doyle, A. S. Broad, C. Kiemle, and
H. Volkert, 2002: Mountain waves over Mont Blanc: In-
fluence of a stagnant boundary layer. J. Atmos. Sci., 59,
2073–2092.
Snively, J. B., and V. P. Pasko, 2003: Breaking of thunderstorm-
generated gravity waves as a source of short-period ducted
waves at mesopause altitudes. Geophys. Res. Lett., 30, 2254,
doi:10.1029/2003GL018436.
Stone, P. H., 1966: On nongeostrophic baroclinic stability. J. Atmos.
Sci., 23, 390–400.
Trier, S. B., and R. D. Sharman, 2009: Convection-permitting
simulations of the environment supporting widespread tur-
bulence within the upper-level outflow of a mesoscale con-
vection system. Mon. Wea. Rev., 137, 1972–1990.
——, ——, R. G. Fovell, and R. G. Frehlich, 2010: Numerical
simulation of radial cloud bands within the upper-level out-
flow of an observed mesoscale convective system. J. Atmos.
Sci., 67, 2990–2999.
Uccellini, L. W., P. J. Kocin, R. A. Petersen, C. H. Wash, and
K. F. Brill, 1984: The Presidents’ Day cyclone of 18–19 February
1979: Synoptic overview and analysis of the subtropical jet
streak influencing the pre-cyclogenetic period. Mon. Wea. Rev.,
112, 31–54.
VanZandt, T. E., and D. C. Fritts, 1989: A theory of enhanced
saturation of the gravity wave spectrum due to increases in
atmospheric stability. Pure Appl. Geophys., 130, 399–420.
Wurtele, M. G., R. D. Sharman, and A. Datta, 1996: Atmospheric
lee waves. Annu. Rev. Fluid Mech., 28, 429–476.
Zhou, X., J. R. Holton, and G. L. Mullendore, 2002: Forcing
of secondary waves by breaking of gravity waves in the
mesosphere. J. Geophys. Res., 107, 4058, doi:10.1029/
2001JD001204.