The Nocturnal Low-Level Jet
No unique definition for the term
'low-level jet' exists in the literature. This is mainly because of the fact
that low-level wind maxima in the boundary layer occur in a number of rather
different situations....Helmut Kraus, et al. in Boundary- Layer Meteorology, Page 187, Vol. 31,
1985.
1. Introduction.
A nocturnal low-level jet
(LLJ) is shown in Figure 1 by a wind-speed profile. It was
measured with a pibal before sunrise during a hot-air balloon competition
at Battle Creek, MI, in June 1989. The profile, a height distribution of
speed, shows a maximum of about 26 knots at 350 feet. The wind speed increases,
from less than 5 knots near the surface, to the 350-foot maximum and then
decreases to about 13 knots near 1150 feet.
A low-level jet has several characteristics significant for hot-air ballooning. First is the wind-speed shear. Between the surface and 350 feet in this example there is a shear of 7 knots per 100 vertical feet. It is obvious that if the height of the speed maximum, i.e., the jet, were lower or its speed greater, or both, the shear would be larger. Some have been reported as high as 10 knots per 100 feet. A large shear, of course, can distort the envelope shape and weaken or eliminate lift.
Figure 2 shows
another important characteristic, the corresponding wind direction changes.
Below the speed maximum there is veering, in this case, of
about 30 degrees, characteristic of northern hemisphere wind direction change
in a stable layer near the surface where the speed increases with height.
(See Turning with the Wind (TWTW), the second tutorial in this series.)
Above the maximum, the wind backs almost 20 degrees while the speed
decreases to what it was at 100 feet.
Good steering below the speed
maximum is related to the existence of the nocturnal ground-based inversion,
a condition for nocturnal LLJ formation. As described in Flying Times
and Windy Times (FTAWT), the first tutorial in this series, an
inversion layer is a stable layer with minimal turbulent mixing so that the
height variation of wind direction is maintained. Above this maximum the
wind backs almost enough to recover one's original ground position. This
layer, however, may exist in varying strengths for a number of reasons including
the winds-aloft pattern.
What neither figure shows
is the important fact that nocturnal jets form and decay at the times most
popular for hot-air balloon activities. Formation begins before sunset as
decreasing heat from the sun decreases daytime instability and turbulent
mixing. The downward mixing of high-speed turbulent air is reduced and the
air's momentum being transferred to the ground during the earlier afternoon
is now redirected forward, forming the LLJ. If it is not too cloudy, an inversion
forms at the ground and more effectively reduces momentum transfer to the
ground, causing the jet to increase during the evening. In the hours before
sunset, flying may be influenced by a range of conditions as the jet forms.
In mid-latitudes the jet
speed typically reaches its maximum an hour or two after midnight and then
decays. After sunrise as the sun heats the ground, it destroys the ground-based
inversion. As the inversion layer decays upward from the ground, it is replaced
by a much less-stable layer. Then turbulence can mix the air so that wind
speed increases at the ground and helps decrease the jet speed. This is the
time when there may be a rapid increase in ground-level wind speed, a special
concern for hot-air balloon pilots who have taken taken off in nearly calm
winds at sunrise.
Not all LLJ's, however, have
such uncomplicated characteristics. One important feature is their tendency
to be controlled partially by an inertial oscillation, as briefly described
below. Another characteristic of low-level jets is one that belies the name
jet. They are actually more like sheets than jets. They may extend
horizontally great distances, both crosswind and along wind, perhaps more
than a hundred miles at times, in comparison to their thicknesses of a few
hundred feet.
The nocturnal low-level
jet described above, and called simply an LLJ, is actually a member of a
subclass of low-level jets found at various times of day in many different
parts of the world. They are caused by a variety of processes, but the term
is usually applied only to those found in the atmospheric boundary layer,
i.e., within a few thousand feet above the ground. Most nocturnal
LLJ's are caused by diurnal changes in the vertical temperature structure
near the ground, due either to sloping terrain or by heat exchanges at the
surface. The LLJ described above comes under the category of those caused
by day-night differences in vertical temperature structure over land without
the influence of sloping terrain. Unless otherwise specified
in the context, this tutorial describes only the latter, for it appears
to be the LLJ phenomenon most frequently encountered by hot-air balloonists.
The definition of a low-level
jet that appears in meteorological literature depends on the purpose of
an investigation, the available data for analysis, or a type of model being
constructed. A definition on the basis of jet speed would be difficult because
there appear to be no natural categories or no limiting conditions applicable
for the wide range of pilot experience and skill.
Most wind-speed statistics
show a nocturnal maximum a few hundred feet above the ground. For example. Figure 2 in FTAWT, shows year-long averages of wind
speeds measured on a television tower near Oklahoma City. Above 300 feet the
average speed increased systematically from an hour or two before sunset to
nearly midnight. At the same time, for most of that period, the wind below
100 feet decreased. An average may be composed of only cases with (a) no changes
and (b) large changes but this does not appear to be the case for diurnal
wind speed changes.
More important than definition
limits is examination of the various conditions observed and the related
boundary-layer factors significant for LLJ development, intensity, and decay.
The examination here consists of the following parts:
1. Introduction
2. The Temperature Profile and the Heat Balance at the Ground
3. Jet Formation
4. Jet Decay
5. Jet Climatology
6. Summary
7. References
In Parts 2, 3, and 4 significant aspects of LLJ formation and decay are described by a one-day example of the processes involved. After a description of heat exchanges at the ground (Section 2), a jet formation late in the day is described and then the decay of a jet formed on the day before is described. These examples were chosen mainly because of the existence of measurements suitable to illustrate some of the important factors.
There are 16 figures. All
consist of graphs prepared especially to describe LLJ's , conditions of
their formation and of decay, and their observed occurrences in special investigations.
Except for Figures 1, 2 and 14, all data for these graphs were taken from
research papers or reports of field investigations. Data for Figures 1 and
2 were obtained from a routine pibal sounding at a hot-air balloon contest,
those for Figure 14 from a simple calculation.
1. The Temperature Profile and the Heat Balance at the Ground
As briefly described above, the formation of a low-level jet begins in the late afternoon and may continue throughout the evening. Formation processes are described in more detail in the following sections, with emphasis on changes in turbulent mixing as revealed by changes in the vertical distribution of average air temperature near the ground. Measurements of vertical distributions of temperature (temperature profiles) are relatively easy to make and, also, their significant characteristics may often be estimated from cloudiness, wind speed and ground conditions. In this way they are useful for anticipating transition-time wind changes.
Figure 3. Major
heat exchange components at the ground on a clear (24-hour) August day in
Nebraska. Transfers to the ground surface are shown as positive, those
away as negative. Data from References 1 and 2.
Figure 3 shows the important heat exchange components at the ground for the day of the LLJ formation and decay described in the following two sections.
In the absence of clouds and with a steady overall flow pattern, the near-ground temperature profile, and consequently the associated turbulent mixing, is largely determined by the heat exchanges between the atmosphere and the ground.
The measurement location
was selected for its simplicity. Covered with short prairie grass, it was
flat, level, and had no trees or buildings to obstruct the flow. The day
chosen for this example was cloudless and the wind above the boundary layer
at 2-3000 feet AGL increased from about 15 knots to 30 knots during the course
of the 24-hour day. This example of ground-surface heat exchanges and the
formation and decay of associated low-level jets was chosen mainly because
of the existence of measurements suitable to illustrate some of the important
factors involved. The major heat exchanges are listed here with brief descriptions
how they were determined.
1.) Heat from the sun, of
course, is the most important. Its pattern on this day is characteristic
of cloudless conditions. It was measured with a sensor mounted 7 feet above
ground out of range of shadows from other instruments and their supports.
2.) Net (thermal) radiation
was measured one meter above the ground. It is the sum of heat from the
sun and infra-red radiation from the atmosphere and any clouds, reduced by
solar heat reflected by the ground, infra-red radiation emitted by the ground
(and vegetation) and a very small amount of reflected cloud and atmosphere
infra-red radiation. During early afternoon on this day, infra-red radiation
from the atmosphere was more than one-third the total from the sun. The ground
reflected about one-fourth of the total solar heat received.
3.) Heat conducted into the
soil, often the smallest of the components, was especially small on this
day because of the dryness of the soil. It was calculated from soil temperature
and direct soil heat conduction measurements.
4.) Evaporation from the
soil and transpiring plants can be a large heat exchange component because
of the heat required to change liquid water to gaseous water. Furthermore,
evaporation consumes heat at the expense of warming the air and the ground.
On this day the evaporative heat loss was relatively small because of the
extremely dry soil. There had been no rain for more than two weeks and the
temperature was above 90 degrees F. on each of the six previous days.
5.) Heat gained by the air during the presence of solar input, and lost during its absence, is achieved largely by direct contact with the ground and/or its cover of vegetation. That's because clear dry air is essentially transparent to solar heat and nearly so to infra-red heat. The combination of the heat exchange components determines the ground's temperature which in turn determines the temperature of the air next to it. With the daytime heat source at the ground, instability develops. At night heat loss is at the ground so stability results.
2. LLJ Formation
Sunset transition time profiles of wind and temperature are shown in FAWT, Figure 9, and are repeated here as Figure 4 and Figure 5 for easy reference. The temperature profiles show the important feature of decrease in stability following the 4:35 PM profile, from being super-adiabatic below 328 feet to an inversion up to about 1500 feet by 2:35 AM the following morning. During this time the wind profile changed from a nearly constant value above 328 feet to a profile with a pronounced jet of about 45 knots at 1312 feet by 2:35 AM.
Each of the following three
figures has the same graph (in red) of the temperature difference between
the heights of 0.3 and 21 feet from 2:30 to 8:30 PM for the day represented
in Figure 3. The temperature scale, T (0.3ft) - T (21ft), is on the
left, showing that the difference ranged from about 8 degrees to -2 degrees
F. In other words, the profile, initially very unstable, changed uniformly
from super-adiabatic to a stable layer as an inversion by 8:30 PM. The dry
adiabatic lapse rate (DALR) is about 0.6 degrees F for the vertical difference
of nearly 21 feet. It is indicated by the light blue horizontal line. Note
that the cross-over from unstable to stable took place an hour before sunset.
Figure 6 shows the temperature difference along
with solar heat, scaled in watts per square meter on the right. The close
correspondence of decreasing instability with decreasing solar heat illustrates
the dominance of the sun in the heat balance at the ground when there are
no clouds.
Figure 7 , with the same temperature graph,
shows four measurements of the horizontal shearing force at the ground during
this period. The shearing force, also called the shear stress, is a measure
of the moving air's momentum loss to the ground by frictional drag. It has
decreased mainly because of the increase in stability that inhibited turbulent
mixing, thereby reducing the downward transfer of momentum.
Following the physical law
of the conservation of momentum, the vertically restrained flow of momentum
to the ground was redirected horizontally, forming a low level jet.
Both theory and observations indicate that jet formation is enhanced by vigorous afternoon turbulent transfer of momentum to the ground, followed by its rapid decrease or cessation. Vigorous turbulent transfer, in turn, is enhanced by relatively strong wind with a large temperature decrease with height in the boundary layer. The latter, of course, is enhanced by abundant solar heat, dry soil and vegetation (i.e, little evaporation) and a relatively cool overriding air mass. Apparently LLJ's are less likely to form if afternoon turbulent transfer is weak and/or slow in decay during the sunset transition period.
Important also for enhanced turbulent transfer is roughness of the ground surface. Not mentioned above, this relationship is supported by theory and boundary-layer observations.
There are many other atmospheric
conditions that influence the formation, or absence of an LLJ, both in the
boundary layer and above . Perhaps the most important, for the type of LLJ
considered here, are the wind conditions above the boundary layer. The jet
formation is enhanced by a wind speed decrease with height above the jet maximum,
but a decrease with time during the night. Apparently the opposite in both
cases is true.
4. Jet Decay
As in Section 3, temperature
and wind-speed profiles from FTAWT are reproduced here for easy reference
(Figures 9 &10) while examining details. The 06:35 AM
temperature profile is characterized by a typical nocturnal inversion from
the ground up to 1400 feet. Stability extends up to at least 2500 feet as
can be seen in comparison to the dry adiabatic lapse rate shown by the green
line. By 10:30 the lower layer has become unstable from the ground to above
300 feet, and neutral above. The 08:30 profile shows how instability develops
first at the ground and builds upward following sunrise.
The 06:35 AM wind-speed profile
shows a jet speed maximum at around 900 feet. By 08:35 the speed has decreased
at all levels except near the surface where it has increased to above 15
knots. The height of the jet speed maximum has effectively increased about
400 feet. By 10:35 the wind speed profile shows little change above 300 feet
corresponding with the lack of stability and enhanced turbulent mixing indicated
by the 10:30 temperature profile.
The following three figures are similar to those constructed for the jet formation description above. They are for the sunrise transition period in the morning of the day represented in Figure 3. Each has a graph of temperature difference, temperature at 0.3 feet minus that at 21 feet, shown in red. Graphs of solar heat, shear stress at the ground, and wind speed changes are shown in sequence with them.
Figure 11 reveals
the important fact that the inversion in the layer from 0.3 feet to 21 feet
lasted for 35-40 minutes after sunrise. Dashed line segments between 5:30
and 6:30 AM are estimates of the actual temperature difference behavior.
(Solid lines are used to connect data from regular hourly or two-hourly measurements.)
Note the correspondence of development of instability with the increase in
solar heat.
Figure 13 shows how wind speed at the ground increased, while speed above the jet decreased. In both cases the changes can be associated with turbulent mixing indicated by the increase in instability both above and below the jet maximum.
But they may also be related
to the jet flow pattern's inertial oscillation, the associated wind velocities
of which are influenced by the Coriolis effect. The latter, described in
TWTW, is essentially a correction applied to the laws of physics in order
to analyze atmospheric motion in relation to the turning earth. A theoretical
treatment of the low-level jet as an inertial oscillation, in effect unattached
from the earth, but viewed relative to it, shows that it would have a (return)
period depending on the latitude.
Figure 14 is a graph of the dependency. It varies
from 12 hours at the poles to 70 hours and more at low latitudes. At the
latitude of the above examples, vis. 42.5 degrees, the period is about
18 hours. This means the jet would return to its starting conditions in 18
hours if there were no other influences. A starting time may be taken as
6 PM so that its cycle would be completed by noon the following day. Without
clouds, as in the above example, turbulent mixing developed between the jet
maximum and the ground would have minimized, if not eradicated it well before
noon.
If the same, or similar, atmospheric conditions were repeated it would begin regeneration by 5 or 6 PM, starting a new cycle. In this case the inertial period could be effective for only three quarters of its cycle. The common observation of a jet maximum an hour or two after midnight, however, may be the result of its inertial oscillation.
Influences of inertia oscillations in the development and decay of LLJ's have not been closely examined. For the above decay example it may be that the decrease in wind speed near and above the jet maximum between 0635 and 0835 AM (Figure 10) was partially due to the inertial decay phase. But as noted above, turbulent mixing of the higher jet speed upward was also at work. The accompanying temperature profile (Figure 9) indicates decreasing stability during this period.
Mixing the momentum upward
could act to diminish the jet speed, decrease the shear between the ground
and the maximum speed, and reduce speed increase at the ground. The inertial
change, on the other hand, if it happens to be in the proper phase, could
enhance the ground level speed increase after sunrise.
5. LLJ Climatology
Reliable statistics of geographic
distributions and frequencies of occurrence are only slowly emerging after
the first full recognition of nocturnal LLJ's, nearly a half century ago.
At least part of the cause of the slow accumulation of data was lack of
appreciation of their low heights and their short periods of formation and
decay. The standard rawinsonde balloon-born sensors rise at about 1000 feet
a minute with the first measurement above ground at about 500 feet, the second
at twice that. The network of rawinsonde stations, about two per state in
the U.S.A. operated only twice daily, cannot provide reliable statistics
of LLJ occurrences. Nonetheless, climatologies of LLJ's have been attempted
from standard rawinsonde data. They have been helpful, but only in a general
way, and for limited application.
Without at least a few years
of LLJ network 
data having adequate height
and time
resolution, usefulstatistics
can be derived
only from special
investigations. One of these
is described in Reference 4. Data were analyzed for nearly a two-year period
(April 1994 to March 1996) from rawinsondes launched five to eight times
a day. The location was at 36.6 deg. N and 97.5 W at the Kansas-Oklahoma
state border. It is near the center of maximum occurrence of LLJ's (east
of the U.S. continental divide) determined by an earlier investigation that
made use of data from standard rawinsonde ascents.
Measurements were made by tracking balloon-born instruments with radio signals from LORAN, the Long Range Navigation system for ships and aircraft. A minimum of three LORAN stations was used to obtain time differences for velocity calculations. Instrument height was calculated from an on-board pressure sensor. This is a considerable improvement over the standard method using radio direction finding equipment and an assumed balloon lift rate. The heights of the lowest wind data obtained were reported to be less than 328 feet for 83% of the observations.
Unfortunately most of the results of this investigation are not sorted in a way to distinguish clearly between nocturnal LLJ's and others. In addition, the statistics of occurrence are based on the total number of soundings for the two-year period, not on the number of days during which soundings were made. For example, the distribution of 1359 jet speed maximums, as determined f rom 46% of the total of 2954 soundings made in the 724 days, was given as follows:
19 Knots & Above...............46.0 % of all soundings
23 " 34.5 "
31 " 18.1 "
38 "
8.7 "
The distribution of jet heights
is shown in Figure 15 at 328-foot (100 meter) intervals. As
can be seen, the maximum occurrence of 17.5 % is at about 1300 feet while
50% of the total were at 1600 feet or below.
The number of soundings taken
in the two year period varied with both time of day and month of year. There
were more than 200 soundings in each month except June in which there were
only 106. During July, August and October there were more than 300 soundings
each with October having the most at 328.
The distribution of soundings by time-of-day is shown in Figure 16. The large variations of the number of soundings during the course of a day and months a year make statistics based on the number of soundings
less meaningful for balloonists
than if they 
had been based on
days and time of day.
Three and a half years after the data described in Reference 4 were obtained another field experiment was conducted near the same location. An intensive investigation of nighttime processes in the boundary layer, it lasted for only the month of October 1999. The nocturnal low-level jet measurements are described in detail in Reference 5.
Data summarized here were obtained with one of a series of recently developed ground-based radar-like pulsed laser tracking systems, known as a High-Resolution Doppler Lidar. To determine the wind it measures the radial velocities of aerosols and is reported to have a time resolution of 1 minute or less and a vertical resolution of 32.8 feet, or less. Other measurement techniques included radars, mini sodars (devices that measure air motion by the Doppler shift of sound waves back-scattered by natural variations in atmospheric density), rawinsondes using GPS, and tower-based sensors.
The Doppler lidar measurements produced LLJ heights significantly lower than those shown in Figure 15, as summarized here:
1. 40% of the speed maxima were below 328 feet;
2. 67% of the speed maxima were below 460 feet;
3. 58% of the speed maxima were between about 14 and 21 knots.
The authors acknowledged that their lidar measurements produced lower LLJ heights than most other measurements in the Great Plains. They suggested that some of the difference could be due to LLJ definitions invoked for analysis reasons but , not surprisingly, emphasized resolution capabilities of different measurement systems. Specifically they question the validity of data that lacked adequate wind measurements below the indicated LLJ heights at or below 350 feet AGL.
In a separate, more general
report of the October 1999 boundary-layer investigation (Reference 6), an
example of a LLJ measured by a rawinsonde is described in detail. Its maximum
speed of about 36 knots is at 400 feet, giving a maximum sheer of nearly 10
knots per 100 feet.
The authors of Reference
5 promise an analysis of LLJ's during morning and evening transition periods
in a future report. One can hope, also, for a careful analysis of various
atmospheric conditions of October 1999 at this Oklahoma-Kansas location.
A comparison with other places and relevant conditions would help establish
the degree of generality of these findings.
6. Summary
Nocturnal low-level jets over level and homogeneous terrain begin forming in the late afternoon because of diurnal changes in wind and turbulence. The latter are caused by changes in components of the heat balance at the earth's surface. The connection between the two changing conditions is shown by changes in vertical air temperature structure indicating diurnal changes in stability. These relationships are most prominent in the absence of clouds and of height and time changes of wind above the atmospheric boundary layer.
As heat recieved from the sun decreases in the late afternoon, temperature of the ground surface cools relative to the air immediately above it. As the cooling progresses, the layer next to the ground, now an inversion and stable, grows in depth. It restricts shear-layer turbulence which earlier has been enhanced by midday heating that made the ground warmer than the air. Especially if the wind is relatively strong, before reduced solar heating, turbulent mixing easily brings air momentum to the ground where it is lost to friction. Development of the stable inversion layer inhibits the downward momentum transport causing air at the height near the inversion top to accelerate forward. This forms the low-level jet.
Many observations show that the maximum speed of such a jet occurs an hour or so after midnight. After that, depending on wind and temperature conditions, it begins to decrease. After sunrise, however, heat balance at the ground changes so that the stable inversion layer is diminished or destroyed and turbulence can mix some of the jet speed to lower layers near the ground.
This sequence of conditions may be modified by the tendency of the jet to be controlled by its own inertial oscillation. The Coriolis effect causes the period of the latter to depend on latitude. Its maximum is 12 hours at the poles and increases to about 20 hours at mid-latitudes, and more than 70 hours near the equator. Daytime heating usually destroys the jet so that an inertial period of much more than 12 hours cannot be maintained. The oscillation, however, may be responsible for observations of jet maxima occurring shortly after midnight.
The rawinsonde upper-air measurement system, in use during World War II and until recent times, was incapable of both time and height resolution required to establish a meaningful climatology of low-level jets. The sensors were lifted by helium-filled balloons at about 1000 feet a minute and released only twice a day from two locations in each of the United States. Data from special field investigations show that jet maxima are generally nearer the ground than indicated by the standard rawinsonde network.
Data from a two-year investigation with significantly improved rawinsonde systems, and measurements made five to eight times a day, showed 50% of the heights at 1600 feet or below. Speeds of jet maxima were 19 knots or above on 46% of the soundings. Measurements were made in Kansas near the Oklahoma border previously determined to be a region of maximum jet occurrence in the United States east of the Rocky Mountains.
A later month-long investigation
(October 1999), near the same location but with a sounding system based
on a Doppler laser technique, found 67% of the soundings had jet maxima at
or below 460 feet. 58 % of the soundings had jet maxima between 14 and 21
knots. An example of a well-developed jet observed in this investigation
had a maximum speed of 36 knots at 400 feet, giving sheer of nearly 10 knots
in 100 vertical feet.
7. References
1. Thornthwaite, C. W., et al. Ed., 1953: Summary of Observations Made at O'Neill, Nebraska, Publications in Climatology, The Johns Hopkins University, Laboratory of Climatology, Seabrook, N.J.
2. Portman, Donald J., 1988: Influence of Evapotranspiration Rates on the Development of Thermals, Technical Soaring, 12, 22-29.
3. Lettau, Heinz H. and Ben Davidson, Ed., 1957: Exploring the Atmosphere's First Mile, Volume II, Site Description and Data Evaluation, Pergamon Press.
4. Whiteman, C. David, et al., 1997: Low-Level Jet Climatology from Enhanced Rawinsonde Observations at a Site in the Southern Great Plains, J. Appl. Meteor., 36, 1363-1376.
5. Banta, R. M., et al., 2002: Nocturnal Low-Level Jet Characteristics over Kansas during CASES-99, Bound.-Layer Meteor., 105, 21-52.
6. Poulos, Gregory S., et al., 2002: CASES-99: A Comprehensive Investigation of the Stable Nocturnal Boundary Layer, Bull. Amer. Meteor. Soc. 83, 555-581.
--------------
Back to: FLYING TIMES AND
WINDY TIMES
--------------