An Introduction to Space Instrumentation,
Edited by K. Oyama and C. Z. Cheng, 1–11.
In-situ density measurements in the mesosphere/lower thermosphere region
with the TOTAL and CONE instruments
Boris Strelnikov, Markus Rapp, and Franz-Josef Lübken
Leibniz Institute of Atmospheric Physics e.V. at the Rostock University (IAP)
Schloss-Str. 6, 18225 Kühlungsborn, Germany
The ionization gauges TOTAL and its improved version CONE are designed to measure neutral air density
on a sounding rocket at very high spatial resolution and precision. Time constant of the CONE instrument
is 0.3 ms which results for a typical rocket velocity of 1000 m/s in an altitude resolution of ∼30 cm. The
very high precision of the density measurements allows to catch density fluctuations of 0.05% and, therefore,
derive power spectrum that covers both inertial and viscous subranges of turbulence spectrum. This, in turn,
makes it possible to precisely derive turbulence energy dissipation rate using spectral model technique. Absolute
laboratory calibrations of the ionization gauges allow to derive absolute density. Because of the pronounced spin
modulation of the measurements, the density profile has to be smoothed, which reduces the altitude resolution
of the absolute density measurements to ∼200 m. Assuming hydrostatic equilibrium, the density profile can be
integrated yielding neutral temperature profile with ∼200 m altitude resolution and precision of ∼4 K.
Key words: In-situ, density, temperature, mesosphere, CONE
1.
Aim of the Instruments (Scientific Motivation) ticles whose genesis crucially depends on atmospheric water vapor content and temperature. Indeed, global satellite
observations going back to 1979 appear to indicate trends
in both NLC brightness and occurrence rate (DeLand et
al., 2007; Shettle et al., 2009). Concerning NLC, detailed
knowledge of the temperature profile seems to be the deciding factor in clarifying their origin. PMSE, in turn, are
further related to MLT turbulence that creates structures in
electron densities which produce radar echoes.
In summary, to study MLT dynamics or PMSE it is important to measure turbulence at those heights. To investigate NLC and PMSE it is necessary to measure temperatures with high resolution. In this chapter we describe an in
situ technique to measure densities and temperatures of neutral air as well as turbulence energy dissipation (or heating)
rate in the MLT region with high precision and high altitude
resolution. The only way to conduct in situ measurements
in the MLT region is to use sounding rockets. This in turn
imposes specific requirements for technical realization of
the instruments.
Since 1990, a total of 24 sounding rockets have been
launched by Bonn University carrying ionization gauges
to measure neutral density and its small scale fluctuations
to determine turbulent parameters. Assuming hydrostatic
equilibrium, the density profile can be converted to a temperature profile with an altitude resolution of ≈200 m. This
is by far the highest resolution available at these altitudes.
1.1 Existing in situ techniques
Regarding in situ methods, density and temperature measurements in the MLT region have efficiently been done
with the Falling Sphere (FS) technique (e.g., Meyer, 1988;
Schmidlin, 1991; Schmidlin et al., 1991; Lübken, 1999;
Müllemann, 2004). This method is based on measurements
of deceleration of a sphere ejected from a sounding rocket
at heights between 80 and 110 km which is tracked by a
The mesosphere/lower thermosphere (MLT) region
(roughly from 50 to 120 km) is of great scientific interest,
since it is coupled to the atmospheric layers above and below. On the one hand, dynamical processes propagating upward from the troposphere reach mesospheric heights and
influence MLT dynamics. On the other hand, solar radiation
penetrates down to MLT altitudes and also influences temperature, dynamics, and composition of this region. Also,
through the circulation processes the MLT is dynamically
coupled to other atmospheric layers. For a text book introducing the physical and chemical concept of this altitude
region see Brasseur and Solomon (2005).
Different scale processes from centimeters up to thousands of kilometers are ultimately interconnected. This
chapter describes a technique to measure small-scale structures with characteristic length scales from centimeters to
some kilometers. The small-scale processes that create such
structures play an essential role in the atmosphere. The behavior of the entire atmospheric system can only be understood if accounting for small-scale processes. For a detailed
textbook on atmospheric dynamics see Holton (2004). The
small-scale structures appear as a result of dynamical processes like turbulence or wave breakdown of e.g., gravity
waves or plasma waves.
Special atmospheric conditions, in particular temperature
structure, at polar MLT region give rise to striking phenomena such as noctilucent clouds (NLC) and polar mesospheric summer echoes (PMSE). These phenomena are not
only interesting in their own, but they are also important indicators of anthropogenic impacts on our atmosphere. Both
phenomena are closely related to the existence of ice parc TERRAPUB, 2013.
Copyright 1
2
B. STRELNIKOV et al.: IN-SITU MEASUREMENTS IN THE MLT
Fig. 1. Photograph and drawing of the CONE instrument. From left to right on the photo: Complete electronics, sensor mounted on the flange, cover
of the sensor, and part of the cylinder that is rocket skin.
ground-based radar when falling through the atmosphere.
This technique however, has an altitude resolution of several kilometers only, which is not sufficient for investigation
of small-scale MLT phenomena such as NLC or PMSE.
In response to the need of high-resolution density and
temperature measurements, the atmospheric physics group
at Bonn University started the development of an ionization
gauge based instrument called “TOTAL” in the end of the
1980s (Lübken, 1987). The name TOTAL means that the
instrument aims to measure the total number density of the
atmosphere. TOTAL yields measurements of absolute number density, neutral temperature, and turbulence parameters
(Hillert et al., 1994).
Turbulence in the MLT region was also investigated
by measuring positive ion density fluctuations using fixed
biased Langmuir (electrostatic) probes e.g., (Thrane and
Grandal , 1981; Thrane et al., 1987; Blix et al., 1990a,
b; Blix and Thrane, 1991). It was shown that the
plasma−namely, electrons and positive ions−can be affected by non-turbulent processes like enhanced recombination of electrons with cluster ions (Röttger and La Hoz,
1990), the effect of charged ice particles in PMSE (Cho et
al., 1992), or plasma instabilities (Blix et al., 1994).
In 1992, the TOTAL instrument was revised and an
improved ionization gauge CONE (“Combined sensor for
Neutrals and Electrons”) was developed (Giebeler et al.,
1993). The time constant of the new instrument was decreased from 4.5 ms (TOTAL) to 0.3 ms by employing an
open geometry for the instrument’s sensor. Just like TOTAL this instrument was also developed in the atmospheric
physics laboratory of Bonn University, which then moved to
the Leibniz-Institute of Atmospheric Physics. The CONE
was successfully used in more than 30 sounding rocket
flights. Both ionization gauges took data at a frequency of
∼3 kHz resulting in an altitude resolution of ∼0.3 m for
a typical rocket velocity of 1000 m/s, used for turbulence
measurements. However, due to the spin modulation the
data must be averaged resulting in an effective altitude resolution of ∼200 m for the total number density measurement
(Rapp et al., 2001).
1.2 Principle
The technique we describe here utilizes the CONE instrument carried by a sounding rocket. CONE measures
neutral and electron densities at the same time. Basically,
CONE consist of two parts: A fixed biased Langmuir probe
to measure electron density and an ionization gauge for neu-
tral density measurements. The first utilizes a well known
technique widely used on sounding rockets to measure both
electron and ion density with very high altitude resolution
and precision. This probe is described in detail elsewhere
(Blix and Thrane, 1993; Giebeler et al., 1993). The second
part of the CONE instrument, the ionization gauge is described in detail in this chapter. The geophysical parameter
that is directly measured is number-density of neutral air.
However, since the Earth’s atmosphere at MLT heights can
most often be considered to be in hydrostatic equilibrium, it
is straightforward to derive temperature profiles from measured densities (see Section 3.1). Moreover, since smallscale fluctuations of neutral density are good tracers for turbulent motions, the small-scale part of measured densities
is used to derive turbulent parameters. To summarize, the
CONE instrument ultimately yields measurements of 1) absolute neutral densities, 2) neutral temperatures, 3) neutral
air turbulence, and 4) electron densities.
2.
Instrument
A photograph and drawing of the entire CONE instrument are shown in Fig. 1. It includes the sensor mounted on
a flange, a vacuum-tight cover for the sensor, electronics,
and a metal cylinder wich is the housing for the electronics.
The sensor is directly mounted on the electronics to reduce
noise that can be introduced to the measured currents before they reach the electronics. The electronics and sensor
are fixed on the flange which, in turn is mounted on the
cylinder.
The sensor has to be covered by a cap under which a
high vacuum (∼10−8 mbar) is permanently supported by a
small ion getter pump (IGP). This is done because it is necessary to run the instrument on the ground to ensure that
instrumental parameters such as emission current, anode
and shielding voltages, and electrometer’s noise level have
their nominal values and are stable. For turbulence measurements it is especially important to ensure that the emission current from the filament, that is the electron source for
the gauge, is constant. The filament (cathode) used for thermal electron emission in ionization gauges can only survive
under pressures below ∼1 mbar. CONE is switched-on at
about two minutes before lift-off. During the upleg of the
rocket flight at about 60 to 70 km altitude the cap of the
CONE is opened (usually it is pulled-off by the rocket motor wich is separated from the payload at that point). To protect the sensor against mechanical damage the three guiding
B. STRELNIKOV et al.: IN-SITU MEASUREMENTS IN THE MLT
3
Fig. 2. Schematic and photo of the CONE instrument: Grids 1 through 3 constitute the ionization gauge; grids 4 and 5 screen the ionization gauge
against ambient plasma; grid 5 is connected to a sensitive electrometer and is an electrostatic electron probe.
Fig. 3. Principle scheme of measurement by CONE’s ionization gauge.
Ground means rocket potential.
rods are used which, when the sensor is opened, are pulled
out by springs (see Fig. 1).
2.1 Sensor
A schematic and photograph of the CONE sensor is
shown in Fig. 2. The outermost grid of the CONE instrument (marked by number 5 in Fig. 2) is positively biased.
Therefore, it attracts negatively charged atmospheric constituents. Obviously, this grid also works as a shield for
positive ions. The next grid (number 4) is negatively biased and hence shields the ionization gauge against the rest
of the ambient plasma. The innermost cathode, collecting
electrode, and anode (numbers 1, 2, and 3 in Fig. 2, respectively) complete the ionization gauge itself. The outermost
grid of CONE (electron collector) and the inner grid of the
ionization gauge (ion collector) are connected to sensitive
electrometers. The currents measured by these electrometers are therefore proportional to local electron and neutral
number density, respectively.
Fig. 3 shows the principle of operation of the ionization
gauge which is the same for both the TOTAL and CONE
instruments. The cathode is heated by a self-regulated circuit with voltage U H = 2 V. It emits electrons that are then
accelerated by the anode voltage U A which is adjustable in
the range from 60 to 150 V. Accelerated electrons ionize
neutral molecules by collisions and the anode collects the
electrons from the measurement volume. Positive ions are
collected by the ion collector and produce the electrometer
current Ielm which is proportional to the local neutral number density. The cathode voltage U K is adjustable in the
range from 2 to 30 V to regulate the emission current, I H ,
which is measured and automatically regulated to be constant.
All the adjustable parameters are optimized in the laboratory and are verified by comparing the rocket soundings with other measurements (e.g., falling sphere). Table 1
summarizes the most important technical data of the two
ionization gauges (CONE and TOTAL) as used during all
rocket flights. The altitude resolutions given in Table 1 are
only valid for the determination of relative density fluctuations n = (n − n ref )/n ref (see Section 3.2). For absolute
densities and temperatures, the altitude resolution reduces
to ∼200 m because the measured currents have to be averaged over at least 0.2 s to avoid the spin modulation (see
above).
2.2 Electronics
The block-diagram of the entire CONE experiment is
shown in Fig. 4.
This electronics was built by the German company
“von Hoerner-Systems”. It consists of three modules:
Emission current regulator, electrometer, and a power supply for the IGP. The ion collector of the ionization gauge
is connected to the autoranging electrometer amplifier that
yields the neutral density measurements. The electrometer’s amplifier automatically switches between five ranges
changing the sensitivity by a factor of eight. The overall sensitivity ranges from 4 nA to 4 nA×84 = 16.4 µA.
All the measured voltages and currents are internally digitized with 16 bit resolution. The analog-digital converter
has a time resolution of better than 0.1 ms and a noise level
(RMS) of better than 2 pA. Such low noise level and high
speed of operation make it possible to measure tiny density fluctuations of a magnitude of ∼0.05% at spatial scales
down to 1 cm.
2.3 Position on payload
The most favorable position to mount the CONE experiment on a sounding rocket is along its symmetry axis. The
best quality data are obtained by facing CONE in the ram.
It is prescribed by the aerodynamics of the rocket flight be-
4
B. STRELNIKOV et al.: IN-SITU MEASUREMENTS IN THE MLT
Table 1. Technical data of the TOTAL and CONE ionization gauges.
Cathode
Ion collector
Anode
Emission current
Vacuum meter constant
Time constant (10−2 mbar)
Sampling frequency
Altitude resolution (v R =1000 m/s)
TOTAL
0.1×1.0 mm, Uc = +10 V
∅ 6.6 mm, Uic = 0 V
∅ 14.5 mm, Ua = +100 V
16 µA
0.4 mbar−1
4.5 ms
4880 Hz
∼4.5 m
cause the shock front that appears due to high rocket velocity (Ma ≈ 3) can be more precisely simulated and/or
measured in a wind-tunnel for those conditions (see Section 3.1). Most often the CONE experiment was mounted
on the rear deck so that it was in the ram during the descent
of the spin-stabilized sounding rocket.
Fig. 5 shows the ECOMA payload that was launched nine
times in the time period from 2006 to 2010 (e.g. Rapp et
al., 2010). CONE was mounted on the rear deck of the 14inch payload. This implies that the best quality data was
measured on the descending part of the rocket trajectory.
3.
Geophysical-data Reduction
The high precision and small time constant of the CONE
instrument make it possible to detect small-scale density
fluctuations in both species (neutrals and electrons) that
arise due to processes like neutral air turbulence (Lübken
et al., 2002) or plasma instabilities e.g., (Blix et al., 1994).
Making use of an absolute laboratory calibration for the
CONE instrument allows derivation of an absolute neutral
air density altitude-profile (Rapp et al., 2001). Because the
density filed is disturbed due to shock front effect, an aerodynamical correction must be applied to arrive at the absolute ambient density measurements (Rapp et al., 2001).
This, however, does not affect the measurements of the
small-scale density fluctuations, which are used to detect
turbulence. In addition, the height profile of neutral number densities is integrated to yield a temperature profile at
∼200 m altitude resolution and an accuracy of ∼3 K (Rapp
et al., 2001, 2002).
3.1 Absolute density and temperature
To derive absolute densities from the measured electrometer current, one has to apply laboratory calibrations. The
calibrations are performed in a vacuum chamber where simultaneous measurements by both CONE and an absolute
pressure standard, e.g. a baratron, are made at different
pressures. The pressures are converted to number-densities
using the temperature of the calibration chamber applying
the ideal gas equation. Fig. 6 shows an example of a calibration curve.
While the procedure to derive absolute number densities
from the ionization gauge current works well in the laboratory where the gas enters slowly into the measuring volume, densities are modified if the gauge is mounted on a
sounding rocket. These rockets typically move at a speed
of several times the speed of sound. A shock front develops in the front of such payloads leading to a disturbed
CONE
∅ 6.0 mm, Uc = +10 V
∅ 20.0 mm, Uic = 0 V
∅ 30.0 mm, Ua = +85 V
14 µA
1.5 mbar−1
0.3 ms
4880 Hz
0.3 m
density and temperature field around the vehicle. Thus the
density measured by CONE or TOTAL is enhanced relative to the ambient density by a ram-factor fram defined
as n meas = fram · n amb , where n meas is the density measured by the instrument and n amb is the undisturbed atmospheric density. For a general discussion of the influence of
aerodynamics on rocket-borne in situ measurements in the
middle atmosphere the reader is referred to Bird (1988) and
Gumbel (2001) and, specifically for the CONE instrument
to Rapp et al. (2001).
The ram-factors for the CONE were experimentally
determined in a wind tunnel of the Facility for Rarefied, Supersonic, and Reactive Flows at the Laboratoire
d’Aerothermique du CNRS at Meudon (France) by exposing CONE directly to the supersonic jet. However, the experimental results for the ram factors cannot directly be applied to atmospheric flight conditions because the temperatures in the wind tunnel flow field are different from the
atmosphere in the MLT region.
The robust and to date best way to determine the ramfactor for a particular sounding rocket configuration and
atmospheric flight conditions is to use the Direct Simulation Monte Carlo (DSMC) method (Bird, 1988). For the
CONE instrument the DSMC results were compared with
experimental values and a satisfactory agreement was found
(Rapp et al., 2001).
Assuming hydrostatic equilibrium, a measured density
profile converted to a temperature profile:
1
m̄ z
T (z) =
(1)
n(z )g(z )dz
n(z 0 )T (z 0 ) −
n(z)
k B z0
where g(z) is the gravitational acceleration, k B is Boltzmann’s constant, m̄(z) is the mean molecular mass of atmospheric air, and n(z) is the number density at altitude
z. Note that the temperature at the upper altitude limit z 0 ,
T (z 0 ), has to be known and is normally taken from a reference atmosphere like CIRA86 (Fleming et al., 1988) or
MSISE90 (Hedin, 1991). However, as atmospheric density
increases exponentially with decreasing altitude, the error
due to the uncertainty in the T (z 0 ) decreases exponentially
with decreasing altitude.
There are several error sources which contribute to the
total error of the final temperatures from ionization gauge
measurements in the middle atmosphere. First, the error
of the density measurement itself is directly imprinted onto
the temperature data (see Eq. (1)). This error is directly
given by the error of the laboratory calibration. In the lab-
5
Fig. 4. Block-diagram of the CONE experiment.
B. STRELNIKOV et al.: IN-SITU MEASUREMENTS IN THE MLT
6
B. STRELNIKOV et al.: IN-SITU MEASUREMENTS IN THE MLT
Fig. 5. ECOMA payload launched during the years 2006 through 2010.
Regarding the error introduced by the ram factor, it has
to be noted that a constant factor in the ram factor does not
influence the result of the temperatures. This is because
the temperature given by Eq. (1) does not depend on the
absolute density n(z) but only on a relative density profile
n(z)/n 0 . Thus the ram factor introduces an error to the
temperature measurement only if its gradient (d fram /dz) is
erroneous.
Table 2 summarizes the error analysis for temperatures.
In order to clarify which specific error source dominates
the total error at different altitudes, both the error from the
density measurement and the error of T (z 0 ) are given.
3.2 Turbulence
The primary steps of our turbulence detection technique
are the following. The measured current I (t) as a function
of flight time t at altitudes between ∼70 and ∼110 km is
converted to relative density fluctuations, also called residuals, r (t), which are determined as:
r (t) ≡
Fig. 6. CONE electrometer current as a function of the density/pressure
which has been measured by a baratron (upper panel). In addition a
polynomial fit to the current is shown. The deviation between the fit and
the measured data is presented in the lower panel.
oratory, the ionization gauge current is measured simultaneously with the pressure measurement of a high quality,
temperature-stabilized, baratron. Fig. 6 shows the result
of such a calibration measurement. In the upper panel the
CONE current is shown as a function of density/pressure
which was measured with the baratron. To convert measured currents from the sounding rocket flights to atmospheric densities, a polynomial is fitted to the calibration
data. The lower panel of Fig. 6 shows the percental difference between the fit and the data. These residuals are
a direct measure of the error of the atmospheric density
measurement. For the relevant pressure range from 10−4
to 0.1 mbar—equivalent to atmospheric altitudes from 105
to 70 km a “typical” density error is 2%.
At altitudes above 100 km the temperature error is governed by the uncertainty in the initial temperature value.
This uncertainty rapidly diminishes with decreasing altitude
(see Table 2). Thus, assuming an uncertainty of 25 K at
110 km, this error has dropped to 4 K at 100 km, 1.5 K at
95 km and 0.2 K at 90 km. While most of the flights with
either CONE or TOTAL reached an apogee of ∼130 km
the starting point can not be chosen at that altitude because
above 110 km the significant abundance of atomic oxygen
leads to additional uncertainty. Therefore, the upper altitude
limit of the temperature retrieval is chosen to be 110 km below which the abundance of atomic oxygen is negligibly
small.
n(t) − n r e f (t)
I (t) − Ir e f (t)
n
(t) =
=
(2)
<n>
n r e f (t)
Ir e f (t)
where the reference current Ir e f (t) can be derived either by
a polynomial fit to the measured current I (t) or as a running
average of the measured time series I (t) (Blix et al., 1990a;
Lübken, 1992). This conversion procedure is demonstrated
in Fig. 7. The resultant residual fluctuations (shown in the
right panel of Fig. 7) are then used as a tracer for turbulence.
To derive turbulent energy dissipation, ε, from our measurements we use the spectral model method introduced by
Lübken (1992). The spectral model is based on the fact that
power spectral densities of spatial turbulent tracer fluctuations are given by a theoretical function that covers both
the inertial and the viscous part of the spectrum. It turns
out that a break in the spectrum occurs at the transition between these turbulent subranges. This break is given by ε
and ν only (ν is kinematic viscosity determined from background densities and temperatures). The theoretical function is fitted to the measured spectrum which determines the
break. This in turn is used to unambiguously derive ε. To be
able to apply the spectral model method, one needs to have
measurements that resolve both the inertial and viscous subranges of the turbulence energy spectrum. For mesospheric
studies, this means that we need to resolve spatial structures
from some meters to approximately one kilometer (see e.g.,
Lübken, 1992; Lübken et al., 1993; Lübken, 1997).
In the following an example of the spectral model method
is given. First, a power spectrum of the measured residuals is calculated. Either a Fourier spectrum of a wavelet
spectrum is used (Lübken, 1997; Strelnikov et al., 2003).
The advantage of the wavelet technique is that in contrast to
the Fourier analysis, which decomposes time series into a
sum of the infinite in-time harmonic functions, namely sine
and cosine, the wavelet analysis decomposes the signal into
the time-frequency domain, also referred to as the phase
space (e.g. Holschneider, 1993). That is, it allows the detection of the time intervals where detected frequencies appear
in the time series. For the description of the basic principles, mathematical details, and different applications of the
wavelet analysis the reader is referred to e.g., Daubechies
(1992), Foufoula-Georgiou and Kumar (1994), Kumar and
B. STRELNIKOV et al.: IN-SITU MEASUREMENTS IN THE MLT
7
Table 2. Temperature errors as a function of altitude. Both the error due to the uncertainty of the density measurement Tn and the error due to the
initial temperature value T (z 0 ) at the upper end of the temperature profile T0 , and also the total error Ttot are given.
z [km]
110
105
100
95
90
85
80
75
70
Tn [K]
5.2
4.6
3.4
1.7
1.5
1.5
1.5
2.0
2.0
T0 [K]
25
12
4
1.5
0.2
∼0
∼0
∼0
∼0
Ttot [K]
25.5
12.5
5.2
2.3
1.5
1.5
1.5
2.0
2.0
Fig. 7. Left panel: Measured current (black profile) as a dependence of the rocket flight time (lower axis) or altitude (upper axis). The red solid dashed
line is the fitted reference current Ir e f , used in Eq. (2) for the conversion to the residuals. Right panel: Residuals r (t) resulting from the current
conversion (Eq. 2).
Fig. 8. Global wavelet power spectra of two 100 m sub-bins within a 1 km bin taken at the 85 km altitude. The dotted profile shows the Fourier power
spectrum of the entire 1 km bin. The bold grey profile shows the global wavelet spectrum of the entire 1 km bin. The bold red profiles represent the
global wavelet spectra of the 100 m sub-bins. The dashed curves show the fitted spectral model of Heisenberg (1948) to the spectra of the 1 km bin
and to the 100 m sub-bins in black and red, respectively. The vertical dash-dotted lines marks the inner scales, l0H , derived from the fit for the spectra
of 1 km and 100 m in black and red, respectively.
Foufoula-Georgiou (1997), Torrence and Compo (1998). A
theoretical spectrum according to Heisenberg (1948) is fitted to the measured spectrum. In the inertial subrange this
model exhibits the classical k −5/3 power law, and in the viscous subrange this model obeys the k −7 power law. Between these subranges a smooth transition takes place.
The Heisenberg power spectrum looks as follows:
(5/3) sin(π/3) 2 Nϑ
a 1/3 f a
2π v R
ε
(2π f /v R )−5/3
·
1/4 8/3 2
1 + 2π f /v R / 2 · 9.90π ν 3 /ε
W(2π f ) =
(3)
8
B. STRELNIKOV et al.: IN-SITU MEASUREMENTS IN THE MLT
Table 3. List of successful flights of the TOTAL and CONE ionization gauges.
Flight
Campaign
Sensor
Year
Date
Time
Place
DBN01
DBN02
DBN03
DAT13
DAT50
DAT62
DAT73
DAT76
DAT84
NBT05
NAT13
L-T01
L-T06
L-I09
L-T13
L-T17
L-T21
SCT03
SCT06
ECT02
ECT07
ECT12
NLTE-1
MDMI05
MSMI03
DYANA
DYANA
DYANA
DYANA
DYANA
DYANA
DYANA
DYANA
DYANA
NLC91 (salvo B)
NLC91 (salvo A)
METAL
METAL
METAL
METAL
METAL
METAL
SCALE
SCALE
ECHO
ECHO
ECHO
NLTE-98
DROPPS
MIDAS/
SPRING
MIDAS/
SPRING
MIDAS/ SOLSTICE
MIDAS/ SOLSTICE
MIDAS/
MACWAVE
MIDAS/
MACWAVE
MIDAS/
MACWAVE
ROMA2003
ROMA2003
ROMA2003
ECOMA2006
ECOMA2006
ECOMA2007
ECOMA2008
ECOMA2008
ECOMA2010
TOTAL
TOTAL
TOTAL
TOTAL
TOTAL
TOTAL
TOTAL
TOTAL
TOTAL
TOTAL
TOTAL
TOTAL
TOTAL
TOTAL
TOTAL
TOTAL
TOTAL
TOTAL
TOTAL
CONE
CONE
CONE
CONE
CONE
CONE
1990
1990
1990
1990
1990
1990
1990
1990
1990
1991
1991
1991
1991
1991
1991
1991
1991
1993
1993
1994
1994
1994
1998
1999
2000
20.02.1990
06.03.1990
13.03.1990
22.01.1990
25.02.1990
06.03.1990
08.03.1990
09.03.1990
11.03.1990
01.08.1991
09.08.1991
17.09.1991
20.09.1991
20.09.1991
30.09.1991
03.10.1991
03.10.1991
28.07.1993
01.08.1993
28.07.1994
31.07.1994
12.08.1994
03.03.1998
06.07.1999
06.05.2000
04:54:00
05:18:28
04:21:00
11:15:00
19:20:00
02:41:00
22:53:00
00:25:00
20:42:00
01:40:00
23:15:00
23:43:00
20:48:00
22:40:00
20:55:15
22:27:30
0:08:00
22:23:00
01:46:00
22:39:00
00:50:33
00:53:00
22:33:00
00:00:00
17:08:00
Biscarosse (44 ◦ N)
Biscarosse (44 ◦ N)
Biscarosse (44 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Kiruna (68 ◦ N)
Kiruna (68 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
CONE
2000
15.05.2000
00:46:00
Andøya (69 ◦ N)
yes
CONE
2001
17.06.2001
00:05:00
Andøya (69 ◦ N)
yes
CONE
2001
24.06.2001
21:21:15
Andøya (69 ◦ N)
yes
CONE
2002
02.07.2002
01:44:00
Andøya (69 ◦ N)
yes
CONE
2002
05.07.2002
01:10:00
Andøya (69 ◦ N)
no
CONE
2002
05.07.2002
01:42:00
Andøya (69 ◦ N)
yes
CONE
CONE
CONE
CONE
CONE
CONE
CONE
CONE
CONE
2003
2003
2003
2006
2006
2007
2008
2008
2010
01.07.2003
04.07.2003
06.07.2003
08.09.2006
17.09.2006
03.08.2007
30.06.2008
12.07.2008
19.12.2010
09:23:00
08:20:00
08:27:00
22:17:00
21:06:46
23:22:00
13:22:20
10:46:00
03:36:00
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
Andøya (69 ◦ N)
no
no
yes
yes
yes
yes
yes
yes
yes
MSMI05
SOMI05
SOMI11
MMMI12
MMMI24
MMMI25
ROMI01
ROMI02
ROMI03
ECOMA01
ECOMA02
ECOMA03
ECOMA04
ECOMA06
ECOMA09
where is the Gamma function ((5/3) = 0.902745) and
v R is the rocket velocity. f is frequency measured in the
reference frame of the sounding rocket. ε is the turbulent
energy dissipation rate, and Nϑ is the variability (or inhomogeneity) dissipation rate (see Lübken, 1992, 1993, for
more detailed explanation). f a is a normalization factor
(we use f a = 2). The constant a 2 = 1.74 was obtained
from laboratory experiments (see e.g., Lübken, 1992, 1993,
Density
data
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
1997). The bold symbols denote the measured quantities.
Two values ε and Nϑ are the unknowns that characterize
turbulence and are the final output of this technique. For
more details on general turbulence theory see textbooks by
Batchelor (1953), Tatarskii (1971), Tennekes and Lumley
(1972), Frisch (1995), Pope (2000) or specifically, on atmospheric turbulence, Zimmerman and Murphy (1977).
The kinematic viscosity ν is calculated from measure-
B. STRELNIKOV et al.: IN-SITU MEASUREMENTS IN THE MLT
9
Fig. 9. Density and temperature profiles measured by TOTAL and CONE instruments from 1990 to 2010.
analysis resolves the high-frequency (small-scale) part of
the spectra in time (altitude) much better that the lowfrequency (large-scale) part, the viscous subrange of the turbulence spectra changes within shorter altitude ranges. It
is seen that the dissipation sub-range of the 100 m-spectra
changes around the mean power spectrum (grey bold profile in Fig. 8) to the both sides, to the stronger and to the
weaker turbulence (smaller and larger scales). This demonstrates difference in resolution of the old and the new analysis techniques.
In summary, the standard technique that utilizes the
Fourier spectral analysis was applied to derive the turbulence energy dissipation rate up to the year 2001. This technique yields altitude resolution for the ε-values of ∼1 km.
The
improved analysis technique (since 2002) that utilizes
Fig. 10. Energy dissipation rates measured by TOTAL and CONE instruthe wavelet analysis yields effective altitude resolution of
ments from 1990 to 2010.
∼100 m (Strelnikov et al., 2003).
ments of neutral temperature and density as recommended 4. Obtained Results
by the US Standard Atmosphere (Sissenwine et al., 1962):
Table 3 summarizes measurements performed by the
atmospheric
research groups from Bonn University and
3/2
βT
ν = µ/ρ,
µ=
(4) Leibniz-Institute of Atmospheric Physics using the both
T +S
ionization gages TOTAL and CONE. The total listing comwhere µ is the dynamic viscosity, ρ is the mass density prises 19 TOTAL and 20 CONE measurements. All this
of the air, T is temperature, S = 110.4 K is Sutherland’s flights yielded successful turbulence, i.e., relative density
constant, and β = 1.458 · 10−6 kg/(s·m·K1/2 ) is another fluctuations measurements. However, as mentioned above,
for the absolute density measurements a good and stable
experimental constant.
Figure 8 demonstrates the fitting procedure. The entire flight conditions are required. That is why some of the
altitude bin of 85±0.5 km is used for fitting (data in grey flights in Table 3 did not yield reliable density/temperature
and fit in black, same in left and right panel). Furthermore, measurements.
An overview of the measured density and temperature
we have used different altitude bins of 100 m size only
(data: red solid line, fit: red dashed line). Vertical dash- profiles is shown in Fig. 9 for summer, winter, and interseadot lines mark the derived inner scale, that is transition sonal measurements in red, blue, and green, respectively.
between inertial and viscous energy subranges as well as Energy dissipation rates measured in summer for the years
the dissipation rates, ε, are shown in each case (black for up to 1999 were published by Lübken et al. (2002) and winter measurements up to 1994 were published by Lübken
1 km bin, red for 100 m bins).
As can be seen, the global wavelet spectrum of the entire (1997). Those works, used an analysis technique resulting
altitude bin (grey bold profile) gives us the smoothed ver- in 1 km altitude resolution. The entire scope of turbulence
sion of the Fourier power spectrum (Torrence and Compo, measurements up to 2010 is summarized in Fig. 10, also for
1998). After the fitting of a theoretical spectrum (black summer, winter, and interseasonal measurements, respecdashed line) this yields the same turbulent characteristics as tively, and also with 1 km resolution. Additionally, mean
our Fourier technique does. However, because the wavelet profiles smoothed over five points are shown.
10
B. STRELNIKOV et al.: IN-SITU MEASUREMENTS IN THE MLT
The most important results obtained with CONE and TOThe obvious future improvement that is foreseen is
TAL are:
miniaturization. Together with other technical efforts this
can bring new scientific insights in turbulence studies. We
• Invention of spectral model technique for the precise
plan to eject several mini-CONEs from one rocket and
determination of ε (Lübken et al., 1993)
probe turbulence structures in several points and thereby
• First reliable climatology of turbulent energy dissipameasure the three-dimensional distribution of turbulent
tion at polar latitudes (Lübken, 1997)
density fluctuations. We also consider to launch CONE on
• Summer climatology of turbulence and its relation to
small meteorological rockets which are much cheaper than
PMSE (Lübken et al., 2002)
sounding rockets. This would allow to derive more detailed
• Characterization of mesospheric turbulence during the
temperature climatology and a good statistics for MLT turArctic spring transition (Müllemann et al., 2002)
bulence.
• Implications of MLT turbulence for gravity wave dissipation and its connection to the residual circulation
(Rapp et al., 2004)
References
• Improved turbulence analysis technique yielding
Batchelor, G. K., The Theory of Homogeneous Turbulence, The Theory
100 m altitude resolution (Strelnikov et al., 2003)
of Homogeneous Turbulence, Cambridge: Cambridge University Press,
• Increased precision of absolute density and tempera1953.
ture measurements by obtaining ram-correction factors Bird, G. A., Aerodynamic effects on atmospheric composition measurements from rocket vehicles in the thermosphere, Planet. Space Sci., 36,
(Rapp et al., 2001)
921–926, doi:10.1016/0032-0633(88)90099-2, 1988.
• High resolution temperature measurements show that Blix, T. A. and E. Thrane, Middle atmosphere turbulence at high and
small-scale temperature variations play a key role in
middle latitudes, 317, 437–445, 1991.
Blix, T. A. and E. V. Thrane, Noctilucent Clouds and regions with polar
the NLC formation process (Rapp et al., 2002)
mesospheric summer echoes studied by means of rocket-borne electron
• Simultaneous measurements of mesospheric temperaand ion DC-probes, Geophys. Res. Lett., 20, 2303–2306, 1993.
ture inversion layer and turbulence at equatorial lati- Blix, T. A., E. V. Thrane, and Ö. Andreassen, In situ measurements of
tudes (Lehmacher et al., 2006)
fine-scale structure and turbulence in the mesosphere and lower thermosphere by means of electrostatic positive ion probes, J. Geophys. Res.,
• First in situ turbulence measurements at 79◦ N (Strel95(D5), 5533–5548, 1990a.
nikov et al., 2006)
Blix, T. X., E. V. Thrane, D. C. Fritts, U. von Zahn, F.-J. Lübken, W. Hillert,
5.
Summary and Outlook
Regarding technical specifications, the advantages of the
CONE instrument are:
• High sensitivity enables to detect very small density
changes down to 0.05%
• Small time constant 0.3 ms provides ability to achieve
very high spatial resolution on a sounding rocket
Apart from the technical side, CONE is a unique instrument
that is able to measure neutral air density with unprecedented spatial resolution. This is a prerequisite to derive
turbulence parameters applying spectral model technique.
Note, that this technique requires that energy spectrum has
to be resolved in wide range of scales: from the smallest
(∼tens of centimeters) to about one kilometer. Also the altitude resolution of the derived temperature profile is very
high. It is scientifically important (e.g., for gravity wave
studies) to be able to resolve small-scale temperature perturbations at short time scales. CONE is so far the only instrument that does this job well in the height range from 70
to 110 km. However, CONE has also some complications.
• It needs careful treatment during launching and countdown due to high vacuum in the sensor volume, which
is protected by the cap. For example, the internal pump
(IGP) must run permanently, also during the rocket assembling and tests.
• The size and weight of the entire CONE experiment
are relatively large. It can only be launched on a big
sounding rocket.
• Turbulence measurement loses sensitivity in the presence of a near adiabatic laps rate.
S. P. Blood, J. D. Mitchell, G. A. Kokin, and S. V. Pakhomov, Small
scale structure observed in situ during MAC/EPSILON, J. Atmos. Terr.
Phys., 52, 835–854, 1990b.
Blix, T. A., E. V. Thrane, S. Kirkwood, and K. Schlegel, Plasma instabilities in the lower E-region observed during the DYANA campaign, J.
Atmos. Terr. Phys., 56, 1853–1870, 1994.
Brasseur, G. P. and S. Solomon, Aeronomy of the Middle Atmosphere:
Chemistry and Physics of the Stratosphere and Mesosphere, 2005.
Cho, J. Y. N., T. M. Hall, and M. C. Kelley, On the role of charged aerosols
in polar mesosphere summer echoes, J. Geophys. Res., 97, 875–886,
1992.
Daubechies, I., Ten Lectures on Wavelets, CBMS-NSF Regional Conference Series in Applied Mathematics, vol. 61, Society for Industrial and
Applied Mathematics, Philadelphia, 1992.
DeLand, M. T., E. P. Shettle, G. E. Thomas, and J. J. Olivero,
Latitude-dependent long-term variations in polar mesospheric clouds
from SBUV version 3 PMC data, J. Geophys. Res., 112, D10,315,
doi:10.1029/2006JD007857, 2007.
Fleming, E. L., S. Chandra, M. R. Schoeberl, and J. J. Barnett, Monthly
mean global climatology of temperature, wind, geopotential height and
pressure for 0–120 km, National Aeronautics and Space Administration,
Technical Memorandum 100697, Washington, D.C., 1988.
Foufoula-Georgiou, E., and P. Kumar (Eds.), Wavelets in Geophysics, Academic Press, Inc., 1994.
Frisch, U., TURBULENCE: THE LEGACY OF A.N. KOLMOGOROV,
Cambridge University Press, The Pitt Building, Trumpington Street,
Cambridge, CB2 1RP, UK, 1995.
Giebeler, J., F.-J. Lübken, and M. Nägele, CONE - a new sensor for in-situ
observations of neutral and plasma density fluctuations, ESA-SP-355,
311–318, 1993.
Gumbel, J., Aerodynamic influences on atmospheric in situ measurements from sounding rockets, J. Geophys. Res., 106, 10,553–10,564,
doi:10.1029/2000JA900166, 2001.
Hedin, A. E., Extension of the MSIS thermosphere model into the middle
and lower atmosphere, J. Geophys. Res., 96, 1159–1172, 1991.
Heisenberg, W., Zur statistischen Theorie der Turbulenz, Zeitschrift für
Physik, 124, 628–657, 1948.
Hillert, W., F.-J. Lübken, and G. Lehmacher, TOTAL: A rocket-borne
instrument for high resolution measurements of neutral air turbulence
during DYANA, J. Atmos. Terr. Phys., 56, 1835–1852, 1994.
Holschneider, M., Localization properties of wavelet transforms, J. Math.
Phys., 34(7), 3227–3244, 1993.
B. STRELNIKOV et al.: IN-SITU MEASUREMENTS IN THE MLT
Holton, J. R., An introduction to dynamic meteorology, Academic Press,
Amsterdam, 2004.
Kumar, P. and E. Foufoula-Georgiou, Wavelet analysis for geophysical applications, Rev. Geophys., 35, 385–412, doi:10.1029/97RG00427, 1997.
Lehmacher, G. A., C. L. Croskey, J. D. Mitchell, M. Friedrich, F.-J.
Lübken, M. Rapp, E. Kudeki, and D. C. Fritts, Intense turbulence observed above a mesospheric temperature inversion at equatorial latitude,
Geophys. Res. Lett., 33, L08,808, doi:10.1029/2005GL024345, 2006.
Lübken, F.-J., TOTAL: A new instrument to study turbulent parameters
in the mesosphere and lower thermosphere, in ESA SP-270: European
Rocket & Balloon Programmes and Related Research, 215–218, 1987.
Lübken, F.-J., On the extraction of turbulent parameters from atmospheric
density fluctuations, J. Geophys. Res., 97, 20,385–20,395, 1992.
Lübken, F.-J., Experimental results on the role of turbulence for the
heat budget of the upper atmosphere, Master’s thesis, Bonn University,
BONN-IR-93-51, 1993.
Lübken, F.-J., Seasonal variation of turbulent energy dissipation rates at
high latitudes as determined by insitu measurements of neutral density
fluctuations, J. Geophys. Res., 102, 13,441–13,456, 1997.
Lübken, F.-J., Thermal structure of the arctic summer mesosphere, J. Geophys. Res., 104, 9135–9150, doi:10.1029/1999JD900076, 1999.
Lübken, F.-J., W. Hillert, G. Lehmacher, and U. von Zahn, Experiments
revealing small impact of turbulence on the energy budget of the mesosphere and lower thermosphere, J. Geophys. Res., 98, 20,369–20,384,
1993.
Lübken, F.-J., M. Rapp, and P. Hoffmann, Neutral air turbulence and temperatures in the vicinity of polar mesosphere summer echoes, J. Geophys. Res., 107(D15), 4273–4277, doi:10.1029/2001JD000 915, 2002.
Meyer, W., Investigations on large and small scale dynamic processes
in the mesosphere and upper stratosphere by means of wind and density measurements above northern Scandinavia in the framework of the
MAP/WINE Project, Ph.D. thesis, 1988.
Müllemann, A., Temperaturen, Winde und Turbulenz in der polaren Sommermesosphäre, Ph.D. thesis, Leibniz-Institut für Atmospärenphysik,
Kühlungsborn-ISSN 1615-8083, 2004.
Müllemann, A., M. Rapp, F.-J. Lübken, and P. Hoffmann, In
situ measurements of mesospheric turbulence during spring transition of the Arctic mesosphere, Geophys. Res. Lett., 29, 1477,
doi:10.1029/2002GL014841, 2002.
Pope, S. B., Turbulent Flows, Cambridge University Press, The Edinburgh
Building, Cambridge, CB2 2RU, UK, 2000.
Rapp, M., J. Gumbel, and F.-J. Lübken, Absolute density measurements in
the middle atmosphere, Ann. Geophys., 19, 571–580, 2001.
Rapp, M., F.-J. Lübken, A. Müllemann, G. E. Thomas, and E. J.
Jensen, Small scale temperature variations in the vicinity of NLC:
Experimental and model results, J. Geophys. Res., 107(D19), doi:
10.1029/2001JD001,241, 2002.
Rapp, M., B. Strelnikov, A. Müllemann, F.-J. Lübken, and D. C. Fritts,
Turbulence measurements and implications for gravity wave dissipation
during the MaCWAVE/MIDAS rocket program, Geophys. Res. Lett., 31,
L24S07, doi:10.1029/2003GL019325, 2004.
11
Rapp, M., I. Strelnikova, B. Strelnikov, P. Hoffmann, M. Friedrich, J. Gumbel, L. Megner, U.-P. Hoppe, S. Robertson, S. Knappmiller, M. Wolff,
and D. R. Marsh, Rocket-borne in situ measurements of meteor smoke:
Charging properties and implications for seasonal variation, J. Geophys.
Res., 115, D00I16, doi:10.1029/2009JD012725, 2010.
Röttger, J. and C. La Hoz, Characteristics of polar mesosphere summer
echoes (PMSE) observed with the EISCAT 224 MHz radar and possible
explanations of their origin, J. Atmos. Terr. Phys., 52, 893–906, 1990.
Schmidlin, F. J., Derivation and application of temperature corrections for
the United States radiosonde, in Symposium on Meteorological Observations and Instrumentations, 7th, New Orleans, LA, Jan. 14–18, 1991,
Preprints (A92-32051 12-47). Boston, MA, American Meteorological
Society, 227–231, 1991.
Schmidlin, F. J., H. S. Lee, and W. Michel, The inflatable sphere—A technique for the accurate measurement of middle atmosphere temperatures,
J. Geophys. Res., 96, 22,673–22,682, 1991.
Shettle, E. P., M. T. DeLand, G. E. Thomas, and J. J. Olivero, Long
term variations in the frequency of polar mesospheric clouds in the
Northern Hemisphere from SBUV, Geophys. Res. Lett., 36, L02,803,
doi:10.1029/2008GL036048, 2009.
Sissenwine, N., M. Dubin, and H. Wexler, The U. S. Standard Atmosphere,
1962, J. Geophys. Res., 67, 3627–3630, doi:10.1029/JZ067i009p03627,
1962.
Strelnikov, B., M. Rapp, and F.-J. Lübken, A new technique for the analysis of neutral air density fluctuations measured in situ in the middle atmosphere, Geophys. Res. Lett., 30, 2052, doi:10.1029/2003GL018,271,
2003.
Strelnikov, B., M. Rapp, T. A. Blix, N. Engler, J. Höffner, J. Lautenbach, F.-J. Lübken, B. Smiley, and M. Friedrich, In situ observations of small scale neutral and plasma dynamics in the mesosphere/lower thermosphere at 79◦ N., Adv. Space Res., 38, 2388–2393,
doi:10.1016/j.asr.2005.03.097, 2006.
Tatarskii, V. I., The effects of the turbulent atmosphere on wave propagation, Jerusalem: Israel Program for Scientific Translations, 1971.
Tennekes, H. and J. L. Lumley, A First Course in Turbulence, The Massachusetts Institute of Technology, USA, 1972.
Thrane, E. V. and B. Grandal, Observations of fine scale structure in the
mesosphere and lower thermosphere, J. Atmos. Terr. Phys., 43, 179–
189, 1981.
Thrane, E. V., T. A. Blix, C. Hall, T. L. Hansen, U. von Zahn, W. Meyer,
P. Czechowsky, G. Schmidt, H.-U. Widdel, and A. Neumann, Small
scale structure and turbulence in the mesosphere and lower thermosphere at high latitudes in winter, J. Atmos. Terr. Phys., 49, 751–762,
1987.
Torrence, C. and G. P. Compo, A practical guide to wavelet analysis, Bull.
Am. Met. Soc., 79, 61–78, 1998.
Zimmerman, S. P. and E. A. Murphy, Stratospheric and mesospheric turbulence, 35–47, D. Reidel Publ. Comp., Dordrecht, Holland, 1977.
B. Strelnikov (e-mail:
J. Lübken
[email protected]), M. Rapp, and F.-