Four mesosphere–lower thermosphere temperature and turbulence
profiles were obtained in situ within
The structure and dynamics of the mesosphere are largely determined by
atmospheric gravity waves (GWs) propagating from the lower atmosphere
The large variability of the northern winter mesosphere is well known
Modeling of mesospheric turbulence has advanced to multi-scale GW
interactions; an example is the interaction of a small-scale GW with large
scale MIL or with larger GWs
An early attempt at multi-point, in situ turbulence measurements was made by
This paper describes the Mesosphere Turbulence Experiment (MTeX) that
employed two payloads with ionization gauges to obtain four profiles at four
different locations within about 30 min. The launch condition was a MIL
observed by Rayleigh temperature lidar at the launch site. A payload
description and first results have been provided by
The purpose of this paper is to examine the spatial and temporal variability
of mesospheric turbulence in relationship to the static stability of the
background atmosphere. We follow the methods developed for neutral density
measurements in the mesosphere and lower thermosphere
Our paper is organized as follows. The next section describes the experiment with special emphasis on the derivation of density and temperature profiles. Section 3 presents individual profiles of buoyancy frequency, fluctuations and energy dissipation rates, as well as wind profiles, wind shears and Richardson numbers. Section 4 discusses our results in the context of other winter measurements of mesospheric turbulence and multi-scale modeling results. The last section contains our summary and conclusions. The Appendix describes results obtained by a small accelerometer for residual drag measurements.
The experiment was designed to study the spatial distribution and temporal
evolution of mesospheric turbulence in the presence of a MIL. Two pairs of
sounding rockets were launched on 26 January 2015, at 09:13 and 09:14 UT,
and 09:46 and 09:47 UT (00:47 LT), respectively, from Poker Flat Research
Range, Chatanika, Alaska (65.13
The main instrument on MTeX was the ionization gauge of the Combined sensor
for Neutrals and Electrons (CONE)
Immediately after nosecone ejection at 52 km and de-spin, the ACS aligned the payload with the velocity vector anticipated for 95 km, halfway in the upleg science window of 70 to 120 km. The spin rate was adjusted close to 2 Hz. The ACS was turned off during the science window in order not to perturb the in situ measurements with cold gas pulses. The ACS was activated again soon after apogee near 156 km; the payload was flipped over and aligned with the anticipated velocity vector for 95 km on the downleg, halfway in the downleg science window, during which the ACS was turned off again.
The MIST payloads contained TMA canisters for upleg and downleg tracer
releases and were only spin-stabilized. The combination of CONE temperature
and TMA wind profiles allowed the calculation of Richardson numbers as in the
earlier “Turbopause” experiment
Horizontal projection of the four sounding rocket trajectories. The triangles mark altitudes 70 to 120 km in 10 km steps on upleg and downleg for the instrumented payloads (MTeX 46.009 and 46.010), and the dashes mark altitudes 90 to 120 km for the chemical tracer payloads (MIST 41.111 and 41.112). The diamonds mark the apogees.
Figure
The CONE instrument is a spherical hot-cathode ionization gauge designed for
pressures up to
The ion current varies between
Four profiles of ion currents observed during the MTeX flights. Each pair of upleg and downleg profiles was obtained with a single CONE instrument.
The graph clearly shows that the sensitivity of the two CONE ionization gauges is different at lower currents corresponding to altitudes above 100 km, while the variation is very small between 70 and 90 km, indicating similar sensitivity for both gauges. Therefore, we expect similar densities for all four profiles at lower altitudes. During the upleg of 46.010 (orange profile), we observed strong disturbances of the ion current near 75 and 80 km, and associated disturbances of the emission current (not shown).
The cause of these disturbances is not understood, but simultaneous observations from a small, sensitive three-axis accelerometer included in the MTeX payload provide indirect evidence of the the existence of a large mesospheric wind. Here we include the main results of this new diagnostic tool. We have deferred the technical details of the accelerometer analysis and comparison with CONE data to the Appendix.
The accelerometer was mounted close the payload spin axis (
The perturbations in the CONE measurement, only on the upleg of 41.010, were
unexpected and unprecedented for this instrument. Upon close inspection, we
noticed coincident small changes in acceleration of
Ion currents observed during laboratory calibrations of the two CONE
sensors. Irregularities at pressures below
The purpose of the sensor calibration is to correct for the nonlinear
variation between pressure and ion current and to account for differences
between individual instruments. The calibration for the two CONE ionization
gauges is shown in Fig.
After applying the calibration, the densities obtained correspond to what is
measured inside the CONE ionization volume, and these values are larger than
the densities in the free atmosphere due to compression effects in the
supersonic flow
Neutral number densities for four profiles after calibration and ram correction. The four profiles agree now for all altitudes, but display significant variability above 100 km.
The temperature profiles
Finally, we calculate the buoyancy frequency as
Temperature profiles and buoyancy frequencies are discussed in the next section.
Our method to calculate atmospheric densities using calibration data and ram
correction follows the standard procedure
The open geometry of the CONE ionization gauge aids in the observation of
very small neutral density fluctuations (
While the energy dissipation rate
As noted already, MTeX was the first experiment with the CONE instrument
mounted on an actively stabilized payload, and aligned closely to the
velocity vector (angle-of-attack close to zero). The spin rate was
actively reduced to about 2 Hz. It is well known that small asymmetries in
the supersonic flow around the ion gauge lead to modulations of the current
signal at the spin frequency and higher harmonics
The time series to be analyzed are relative fluctuations of the ion current
Example of CONE relative density fluctuations for flight 46.009 upleg.
Two regions of neutral density fluctuations can immediately be recognized
around 71 and 76 km altitude. Note that the level of fluctuations is much
less than 0.01 (1 %). A small spin modulation becomes more prominent above
85 km. The increasing noise above 90 km was caused by interference from the
voltage sweeps of the Langmuir probe on one of the booms
We have used the wavelet method first applied to CONE data by
Example of wavelet spectrum of turbulent fluctuations. The blue curve is a best fit of a theoretical spectrum including the transition from the inertial subrange to the viscous subrange. For details see text.
Individual wavelet spectra were averaged over 100 m intervals. The thick
black line is such an averaged wavelet spectrum for the interval 71.0 to
71.1 km. The blue line is the least-square fit of a Kolmogorov–Heisenberg
spectrum for stationary, homogeneous, isotropic turbulence with slopes
We will discuss temperature, buoyancy frequency and turbulence results in Sect. 3.
Both MTeX launches were closely followed by two MIST payloads for wind and
turbulence measurements in the lower thermosphere. TMA trails were released
on the upleg and downleg between
Temperature profiles (solid lines) derived from densities in Fig.
Figure
Characterizing the large features of the profile, we observe a relatively
warm winter mesosphere up to 80 km, a quasi-adiabatic region between 80 and
88 km, another stable region up to 95 km, followed by a second
quasi-adiabatic region up to the mesopause at 170 K and 102 km. The two
bottommost regions agree well with the Rayleigh lidar temperatures
Detail of temperature profiles in the lower mesosphere. Same data as
in Fig.
Figure
Temperature profile, buoyancy frequency, wavelet spectra of neutral density fluctuations, and turbulent energy dissipation rates for flight 46.009 upleg. For details see text.
Figures
Same as Fig.
The first profile was obtained on the upleg of flight 46.009. Two distinct
layers of turbulence were observed centered around 71 and 75 km,
respectively. Energy dissipation rates in these layers ranged from 0.18 to
6.4 mW kg
Same as Fig.
The second profile was observed on the downleg of flight 46.009, 70 km north
of the upleg profile. The temperature profile is significantly different in
the lower mesosphere and has a broad maximum at 72.5 km associated with a
deep turbulence enhanced layer. The maximum energy dissipation rate is found
at 72.8 km with 36 mW kg
The third profile was obtained on the upleg of flight 46.010. Its location
was very close to the first profile and the measurement occurred 33 min
later. As explained above, we do not have a complete density and temperature
profile due to the anomaly of the CONE sensor, and the interpolated regions
are marked with dashed lines. However, it can be assumed that the temperature
profile did not change dramatically compared to the first profile, as can be
seen below 73 km and also above 82 km where the measurements were
undisturbed. This is also confirmed by the series of Rayleigh lidar profiles.
With this caveat in mind, we did the wavelet analysis of the ion current
fluctuations in the perturbed regions and found that the spectra conformed
with our turbulence model, although the fluctuations were much amplified in
the perturbed regions, as can be seen from the red contours in the spectral
plot. Compared to the 46.009 upleg, the turbulent layer near 70 km had
weakened or moved by advection, although there is still a thin layer present
at 70.5 km with 1.8 mW kg
Same as Fig.
The last profile was obtained on the 46.010 downleg, about 10 km west of the
first downleg. The temperature profile agrees in many details with the other
three profiles, most significantly, the local maximum near 80 km below a
deep layer with low static stability, as just discussed. At the lower
altitudes we find very weak turbulence in several narrow layers near 70,
72.5, and 74 km, mostly associated with stable regions of the atmosphere.
The largest value is 2.3 mW kg
In summary, and perhaps not surprisingly, we observe strong similarities in the large-scale temperature and stratification structure, but great variability in the altitude, thickness, and strength of the fluctuation layers. It is important to note that, common in all four profiles, turbulent spectra are found mostly in the more stratified region below 80 km, while fluctuations and turbulence are largely absent in the well-mixed layer above.
Figure
Horizontal winds derived from chemical trails.
Another presentation of the wind components are the hodographs in
Fig.
Hodograph projections of the horizontal wind. The lowest altitudes start on the right at positive zonal winds. Small open circles are drawn every 1 km and large filled circles every 10 km. The first large circle is at 90 km. The cross marks the origin (zero wind).
The simultaneous measurements of temperatures with the CONE instrument and
winds with the chemical tracer technique allow the calculation of Richardson
numbers as an index for instability
In Figs.
Buoyancy frequency from CONE upleg temperatures, horizontal wind
shears from TMA winds, and Richardson numbers for the first salvo. The red
line in the right panel indicates
The extreme wind shear is also directly visible in the images of the puffed
TMA trails. Figure
Same as the previous figure but for the second salvo.
Below
Photographs of the MIST-1 upleg trail from Toolik Lake
The MTeX experiment was the first time that four profiles of in situ neutral turbulence and background temperature were obtained close together in time and space. While this is still a very small sample of the turbulent flow field, it allows a limited comparison with high-resolution multi-scale gravity wave breaking simulations.
Photographs of the MIST-1
During our experiment, we did not encounter a large MIL of the type set as
the temperature background in the simulation. However, the mesosphere was on
average stable below 80 km and very weakly stable or quasi-adiabatic between
80 and 88 km. Almost all of our turbulence layers were observed in the more
stable region below 80 km. The perturbations in stability due to GW
interactions resemble the individual
Comparing values for the energy dissipation rate, the statistics of winter
turbulence measurements obtained at Andøya
In an earlier experiment from Poker Flat, fluctuation activity was small in
the lower mesosphere, despite a prominent mesospheric inversion layer at 70
to 75 km
Another winter case study was presented by
A recent sounding rocket flight from Andøya was equipped with two CONE
instruments to provide measurements on the upleg and downleg
In both mesospheric turbulence experiments from Alaska (2009 and 2015) we
have observed nearly adiabatic layers in the upper mesosphere accompanied by
overturning events at the bottom or middle of the sodium layer, respectively.
Such structures in the sodium layer have been modeled and are thought to be
associated with large-scale gravity waves, that are overturning, either
partially or fully, but not breaking
A large number of winter measurements confirms that strong isotropic
turbulence is rarely observed above 95 km.
MTeX was the first sounding rocket experiment that obtained four in situ
temperature and neutral turbulence profiles within 33 min in the winter
mesosphere. In this paper we examined the spatial and temporal variability of
mesospheric turbulence in relationship to the static stability of the
background atmosphere. The four temperature profiles showed a high degree of
consistency at large scales. Two relatively stable regions existed between 68
and 82 km and between 88 and 95 km and two nearly unstable regions between
82 and 88 km and again between 95 and 102 km. The temperature structure was
also observed by Rayleigh lidar up to 90 km
Between 85 and 115 km, we obtained simultaneous wind measurements from TMA
tracer trails and were able to derive Richardson numbers as a measure of
dynamical instability. This was the second experiment in which we obtained
Richardson numbers from the combination of ionization gauge temperatures and
TMA winds. While the earlier “Turbopause” experiment was conducted under
geomagnetically quiet conditions, but during significant gravity wave
activity
The stable region between 68 and 82 km did not have a persistent positive
temperature gradient as in major MIL events and as modeled by
The experiment confirmed that the winter mesosphere is highly variable, and
on the day of the experiment, gravity wave activity in the upper stratosphere
and lower mesosphere was lower than normal
The sounding rocket experiment was funded by NASA's
Heliophysics program. In accordance with NASA's data sharing policy, the data
sets are public. The data that support the findings of this study are
available from the corresponding author on reasonable request. SABER
temperature profiles are available and were retrieved from
A small three-axis MEMS (microelectromechanical system) accelerometer (type
Kionix KXR94-2050) with sensitivity 1
Accelerometer drag residual for flight 46.010 upleg
Figure
Next, we show the acceleration component along the
A DSMC of the supersonic flow using the velocity, density, and temperature
conditions for this flight yielded a drag force of 7.2 N at 75 km. We used
the NASA DAC97 package for our simulations
The perturbations in the CONE measurement were unexpected and unprecedented,
and the simultaneous change in acceleration of 0.5 m
While sensitive accelerometers on supersonic free-falling spheres have been
used previously to successfully measure winds, densities and temperatures in
the mesosphere and lower thermosphere
GAL is co-investigator of the MTeX project, performed the sounding rocket data analysis, drafted the manuscript and prepared the figures. RLC is the principal investigator of the MTeX project and provided the lidar data. AB is co-investigator of the MTeX project and provided the accelerometer experiment. MFL is the principal investigator of the MIST project and provided the tracer images and wind profiles. BS provided the ionization gauges and the code for the wavelet based turbulence analysis. All co-authors provided scientific input as a team effort and during the manuscript preparation and review process.
The authors declare that they have no conflict of interest.
This research was supported by NASA grants NNX13AE35G (Embry-Riddle Aeronautical University), NNX13AE26G and NNX14AH45G (Clemson University), and NNX13AE31G (University of Alaska Fairbanks). The CONE sensors were built by Hans-Jürgen Heckl and calibrated by Arthur Szewczyk at the Institute for Atmospheric Physics in Kühlungsborn, Germany. The CONE electronics was designed and built by von Hoerner & Sulger GmbH, Schwetzingen, Germany. We thank NASA Wallops Flight Facility and Poker Flat Research Range for mission and payload design and launch and recovery operations. The topical editor, Petr Pisoft, thanks two anonymous referees for help in evaluating this paper.