Vol 450 | 29 November 2007 | doi:10.1038/nature06239
LETTERS
The structure of Venus’ middle atmosphere and
ionosphere
M. Pätzold1, B. Häusler2, M. K. Bird3, S. Tellmann1, R. Mattei2, S. W. Asmar4, V. Dehant5, W. Eidel2, T. Imamura6,
R. A. Simpson7 & G. L. Tyler7
100
90
–3.5 K km–1
80
Altitude (km)
Mesosphere
–7.6 K km–1
70
Tropopause
60
Troposphere
The atmosphere and ionosphere of Venus have been studied in the
past by spacecraft with remote sensing1–4 or in situ techniques3,4.
These early missions, however, have left us with questions about,
for example, the atmospheric structure in the transition region
from the upper troposphere to the lower mesosphere (50–90 km)
and the remarkably variable structure of the ionosphere.
Observations become increasingly difficult within and below the
global cloud deck (,50 km altitude), where strong absorption
greatly limits the available investigative spectrum to a few infrared
windows and the radio range. Here we report radio-sounding
results from the first Venus Express Radio Science5 (VeRa)
occultation season. We determine the fine structure in temperatures at upper cloud-deck altitudes, detect a distinct day–night
temperature difference in the southern middle atmosphere, and
track day-to-day changes in Venus’ ionosphere.
The atmosphere and ionosphere of Venus were sounded by VeRa
at 2.3 and 8.4 GHz (the S and X bands, respectively) during the first
occultation season from mid-July to the end of August 2006. As seen
from the Earth, the spacecraft disappeared behind the planetary disk
on every 24-hour orbit during this period5. The radio ray path is
refracted according to the local state of the atmosphere, the degree
of refraction being proportional to the electron density (ionosphere)
or the neutral gas density (atmosphere). The VeRa observations are
the first to use two simultaneous one-way downlink frequencies driven by an Ultrastable Oscillator for radio sounding of Venus’ atmosphere. This enables observations during both ingress and egress,
rather than being restricted to ingress (as with the Pioneer Venus
Orbiter in the majority of cases). Forty-two profiles were obtained
from 21 occultation passes during the first occultation season (one
ingress and one egress per pass). These profiles reveal the ionospheric
structure at altitudes of 100–500 km and the neutral atmosphere in
the altitude range 50–90 km. Longitudes span the range from 296u E
to 352u E for the daytime and from 189u E to 251u E for the night-time
profiles. Additional information on the geometry and coverage during the first Venus Express occultation season, which covers the
Venus Express orbits numbered 81–131 (2006 day of year: DOY
192–242), is provided in the Supplementary Information. Details
on the experiment performance5, the extraction of atmospheric profiles6 and results from additional components of the VeRa investigation, including the first bistatic radar experiments, are published
elsewhere.
To derive the temperature profile, however, we need to know the
composition of the atmosphere (96.5% CO2 and 3.5% N2)7 and the
assumption of an upper boundary temperature. Figure 1 shows three
VeRa temperature profiles derived from the same data set taken
–10 K km–1
50
233 K
150
200
250
300
350
400
Temperature (K)
Figure 1 | Temperature profiles from DOY 234, 2006 at latitude 716 N,
derived with three different upper boundary temperature conditions of
170, 200 and 230 K. Regardless of the upper boundary condition, all three
profiles converge to the same temperature distribution below 90 km. The
temperature shows a constant cooling (lapse rate) of 210 K km21 within the
cloud deck below 60 km. The inversion at 62 km marks the tropopause, the
transistion from the troposphere to the mesosphere. On average, the
temperature is isothermal (233 K) within the upper cloud deck up to 75 km,
but displays significant fine structure, a series of small inversions, with
amplitudes larger than the measurement error. Above 75 km, the cooling
rate observed above the transition is 27.6 K km21 up to 80 km altitudes and
then drops to 23.5 K km21 above this height.
1
Rheinisches Institut für Umweltforschung, Abt. Planetenforschung, Universität zu Köln, Aachener Strasse 201–209, 50931 Köln, Germany. 2Institut für Raumfahrttechnik, Universität
der Bundeswehr München, 85577 Neubiberg, Germany. 3Argelander-Institut für Astronomie, Auf dem Hügel 71, 53121 Bonn, Germany. 4Jet Propulsion Laboratory, California Institute
of Technology, Pasadena, California 91109, USA. 5Observatoire Royal de Belgique, 3 Avenue Circulaire, 1180 Brussels, Belgium. 6Institute of Space and Astronautical Science, Japan
Aerospace Exploration Agency, 3-1-1 Yoshinodai, Sagamihara, Japan. 7Space, Telecommunications, and Radioscience Laboratory (MC 9515), Stanford University, Stanford, California
94305-4020, USA.
657
©2007 Nature Publishing Group
LETTERS
NATURE | Vol 450 | 29 November 2007
a
b
DOY 195, lat. = 62.96º
DOY 196, lat. = 65.58º
DOY 200, lat. = 73.10º
DOY 204, lat. = 78.34º
DOY 195, lat. = 1.07º
DOY 196, lat. = –4.73º
DOY 200, lat. = –24.7º
DOY 204, lat. = –42.06º
100
Altitude (km)
90
80
70
60
50
Ingress
150
200
Egress
250
300
Temperature (K)
350
400
150
200
250
300
Temperature (K)
350
400
Figure 2 | Venus temperature height profiles from VeRa radio-sounding
observations. Four temperature profiles are shown from occultation ingress
at southern mid-latitudes (a) and egress at northern high polar latitudes
(b). Reliable measurements of the received radio signal are not available at
altitudes below about 50 km. The profiles are well determined below 90 km.
Temperature inversions are detected at altitudes from 60 to 70 km. This
inversion range (boxed area) occurs within the cloud deck and is more
confined in the mid-latitudes (65–69 km) than at the poles (63–74 km).
during occultation egress on 2006 DOY 234 at northern polar latitudes
with three different upper boundary temperatures. Regardless of the
particular upper boundary temperature selected, all three profiles
converge rapidly into the same temperature profile, providing confidence in the reliability of the temperature determination below
90 km. The sensitivity of VeRa extends the atmospheric profiles to
higher altitudes, thus providing better continuity to the lower boundary of the Venus Express instrument SPICAV’s atmospheric observations8. Figure 1 typically reveals the fine-scale thermal structure
in the transition region from the upper troposphere to the lower
mesosphere within the upper cloud deck. The Venus atmosphere
reaches the one-bar pressure level at an altitude of about 50 km.
Below this height, the radio signals also suffer significant absorption
by the gas and cannot be used to sound the atmosphere at all below the
super-refractive limiting height of about 35 km (ref. 9).
Figure 2 shows examples of temperature profiles derived from
ingress (Fig. 2a) and egress (Fig. 2b) measurements. A decrease in
temperature with increasing height (lapse rate, 210 K km21) is
found below an altitude of 60265 km. Above this level is a region
of thermal inversions—a series of warmings and coolings within the
range of altitudes attributed to the upper cloud deck. The series of
inversions is constrained by two sharp lower and upper boundaries
below 65 km and above 70 km, respectively, where again a clear cooling trend is observed. The distance between these boundaries is larger
by a factor of three in the polar latitudes (Fig. 2b) than in the equatorial to mid-latitudes (Fig. 2a). The lower inversion is more pronounced in the polar latitudes and increases in altitude towards
lower latitude, consistent with a convergence to the profiles seen in
a
Dayside
85
Altitude (km)
80
75
70
65
60
55
–80
–60
–40
–20
0
b
Nightside
85
Altitude (km)
80
75
70
65
60
55
–80
–60
–40
Planetary latitude (deg)
–20
0
Figure 3 | Venus temperature maps derived from VeRa radio occultation
data. Contours (in K) are generated using all available temperature profiles
in the southern latitude range from 0u to 90u for the dayside (a, 50u , solar
zenith angle , 90u) and nightside (b, 90u , solar zenith angle , 115u) of
Venus. The range in altitude is from 55 to 90 km. The daytime temperature
increases by about 30–40 K from the southern pole to the equator at altitudes
below 60 km. This contrasts with the roughly constant night-time
temperature below 60 km. The inferred result is a day–night equatorial
temperature difference that reaches about 40 K at 55 km altitude. The ‘‘cold
collar’’, a temperature depression encircling the pole in both hemispheres15,
is clearly observed on both the dayside and the nightside between 60u and 80u
latitude at 63 km altitude. The collar extends to 40u latitude during the night
in the southern hemisphere.
658
©2007 Nature Publishing Group
LETTERS
NATURE | Vol 450 | 29 November 2007
Fig. 2a. Although, on average, the temperature is roughly isothermal
within these two boundaries at an average neutral scale height of
5.2 km, the series of inversions implies a thermal fine structure within
the upper cloud deck never before seen with the clarity that is best
exemplified in Fig. 1. Similar inversion behaviour was observed with
the Pioneer Venus Orbiter9, but was much more pronounced at polar
latitudes than was observed with Venus Express.
All ingress profiles from the southern hemisphere are combined
together in the contour plots of Fig. 3. Significant differences are found
between the daytime and night-time temperature distributions (a
colour plot of Fig. 3 is provided in the Supplementary Information).
As an example, Fig. 4a shows an electron density profile from the
ionosphere from 2006 DOY 196, observed in the equatorial region at
15:20 local time and a solar zenith angle of 50u. The noise level of
the retrieved ionospheric profiles is ,2 3 109 m23, which is 0.5% of
the peak electron density. The VeRa ionospheric profiles are thus
of the same high quality as those from the most recent Venus mission,
Magellan10. We can readily identify the bottom and main layers, V1
and V2, respectively. Although the primary ionization product is
CO21 generated by photoionization of solar extreme-ultraviolet
and soft X-rays, the main constituent of the V1 and V2 layers is
O21, formed by rapid molecular reactions of CO21 with atomic
oxygen11. The O1 ion becomes important and controls the density
above 180 km potentially in its own layer, V3 (ref. 11). The majority
of electron density profiles display a bulge in the topside between 160
and 180 km that is not documented in theoretical models of ion and
electron production11.
Figure 4b shows a collection of daytime electron density profiles.
Large variations can be seen in the topside density distribution at the
same time that the lower V1 and V2 layers are relatively stable, with
peak densities close to those expected for the solar minimum conditions of summer 2006 (ref. 12). A smaller peak density is observed
for the profile on DOY 212 at high solar zenith angle. The base of
the daytime ionosphere is stable at an altitude of 120 km. The peak
density altitude of V1 at about 140 km marks the altitude where the
soft X-ray intensity has decreased by e21; radiation from the broad
X-ray spectrum penetrating deeper into the atmosphere does not
produce detectable ionization below 120 km.
The ionopause, the distinct boundary between the solar wind
flow and the planetary ionosphere, is present in the altitude range
225–375 km. These are quite typical for the applicable range of solar
zenith angles near the solar activity minimum12,13. The location is
clearly defined by the strong electron density drop into the noise level
towards higher altitudes. The four profiles of Fig. 4b cover a time
interval of two weeks. While the ionopause heights on DOY 196 and
200 are nearly the same, the height is almost 100 km higher on DOY
202. Ten days later the ionopause returned to its previously lower
level. This may imply that the balance between the solar wind
dynamic pressure and the ionospheric plasma pressure was stable
over many days, but then changed significantly from one day to
another. It is remarkable that all profiles display roughly the same
electron density gradient near the ionopause—falling into the noise
over an altitude range of about 30 km.
The ionosphere was also sounded on the Venus nightside (Fig. 5).
We distinguish in this case between profiles from the ‘deep’ night
with solar zenith angles greater than 98u and those close to the terminator (solar zenith angle < 90–98u). Profiles from the deep night
vary in character, sometimes showing significant unstructured
ionization, while at other times no ionization can be detected. This
is consistent with Pioneer Venus Orbiter results during solar
a
b
300
375
DOY 196, SZA = 50°, lat. = –5°
DOY 200, SZA = 56°, lat. = –25°
DOY 202, SZA = 59°, lat. = –34°
DOY 212, SZA = 80°, lat. = –73°
280
325
260
Ionopause
240
Altitude (km)
275
220
V3
200
225
180
175
160
V2
140
125
120
100
10–2
V1
10–1
100
Electron density (1011 m–3)
Figure 4 | Daytime electron density profiles in the ionosphere of Venus.
a, The main features of a Venus ionospheric electron density height profile
derived from the VeRa occultation data. This ingress example of 15 July 2006
(DOY 196), derived from simultaneous Doppler recordings of the occulted
2.3 and 8.4 GHz signals and from the associated differential Doppler data
calculated from both sets of measurements, shows the ionosphere at solar
zenith angle 50u, local time 15:20 and latitude 24.7u. The differential
Doppler data are sensitive only to changes in the ionized medium along the
radio propagation path—in this case, primarily in the Venus ionosphere.
This profile is essentially identical with those derived from the X-band or
S-band data sets alone, providing confidence in the quality of the derived
electron density profiles even from single-frequency Doppler data. Electron
101
75
10–2
100
10–1
Electron density (1011 m–3)
101
density layers are identified as the secondary and main layers V1 and V2,
respectively. A V3 region is formed above 180 km controlled by the O1 ion11.
A bulge in the topside between 160 and 180 km altitude (arrow) is apparent
in the majority of the observed daytime profiles, and is not explained by
models of the Venus ionosphere. The ionopause is located between 250 and
275 km in this example. b, VeRa daytime ionospheric electron density
profiles on four different days. These ingress profiles were derived from the
8.4 GHz Doppler data. Clearly identified are the ionosopheric base at
120 km, the secondary and main layers V1 and V2, respectively, the V3 layer,
a highly variable topside, and an ionopause at altitudes varying between 225
and 375 km. The latitude and solar zenith angle (SZA) associated with each
profile is indicated.
659
©2007 Nature Publishing Group
LETTERS
a
NATURE | Vol 450 | 29 November 2007
densities result from the longer atmospheric absorption lengths
and corresponding reduction in ionizing flux at these high solar
zenith angles.
b
350
300
Received 23 May; accepted 5 September 2007.
1.
Altitude (km)
250
2.
200
3.
150
4.
100
5.
DOY 218, lat. = –84°, SZA = 92.4°
DOY 233, lat. = –23.5°, SZA = 113.0°
50
c
d
350
7.
300
8.
250
Altitude (km)
6.
9.
200
10.
150
11.
100
12.
DOY 234, lat. = –18.8°, SZA = 113.4°
50
DOY 239, lat. = 6.5°, SZA = 113.5°
0.5
13.
Figure 5 | Four electron density profiles from the nightside of Venus. a, An
example from high polar latitudes at 284u and near the terminator at solar
zenith angle solar zenith angle 5 92.4u, when the upper atmosphere is still
illuminated and a structured ionosphere has formed. The peak density at
145 km is about an order of magnitude less than that of the mid-latitude
daytime profiles shown in Fig. 4. b–d, Profiles at various latitudes obtained
for solar zenith angles near 113u and local time 05:00 (that is, unilluminated
up to almost 500 km). Only weak, possibly sporadic, ionization was observed
in some profiles. In contrast to the logarithmic representation of the daytime
profiles in Fig. 4, these night-time profiles are plotted on a linear scale.
14.
0.0 0.1 0.2 0.3 0.4 0.5
Electron density (1011 m–3)
0.0 0.1 0.2 0.3 0.4
Electron density (1011 m–3)
minimum, when ion convection from the dayside ionosphere is
reduced14. Profiles from locations close to the terminator at polar
latitudes show a layered structure similar to that observed during
the day, albeit with much lower peak densities. The reduced
15.
Mariner Stanford Group. Venus: ionosphere and atmosphere as measured by
dual-frequency radio occultation of Mariner V. Science 158, 1678–1683 (1967).
Howard, H. T. et al. Venus: mass, gravity field, atmosphere, and ionosphere as
measured by the Mariner 10 dual-frequency radio system. Science 183, 1297–1301
(1974).
Hunten, D. M., Colin, L., Donahue, T. M. & Moroz, V. I. (eds) Venus (Univ. Arizona
Press, Tucson, 1983).
Bougher, S. W., Hunten, D. M. & Phillips, R. J. (eds) Venus II: Geology, Geophysics,
Atmosphere, and Solar Wind Environment (Univ. Arizona Press, Tucson, 1997).
Häusler, B. et al. Radio science investigations by VeRa onboard the Venus Express
spacecraft. Planet. Space Sci. 54, 1315–1335 (2006).
Fjeldbo, G., Kliore, A. J. & Eshleman, V. R. The neutral atmosphere of Venus as
studied with the Mariner V radio occultation experiments. Astron. J. 76, 123–140
(1971).
von Zahn, U., Kumar, S., Niemann, H. & Prinn, R. in Venus (eds Hunten, D. M.,
Colin, L., Donahue, T. M. & Moroz, V. I.) 299–430 (Univ. Arizona Press, Tucson,
1983).
Bertaux, J. L. et al. A warm layer in Venus’cryosphere and high-altitude
measurements of HF, HCl, H2O and HDO. Nature doi:10.1038/nature05974
(this issue).
Kliore, A. J. & Patel, I. R. Vertical structure of the atmosphere of Venus
from Pioneer Venus Orbiter radio occultations. J. Geophys. Res. 85, 7957–7962
(1980).
Jenkins, J. M., Steffes, P. G., Hinson, D. P., Twicken, J. D. & Tyler, G. L. Radio
occultation studies of the Venus atmosphere with the Magellan spacecraft. 2.
Results from the October 1991 experiments. Icarus 110, 79–93 (1994).
Fox, J. L. & Sung, K. Y. Solar activity variations of the Venus thermosphere/
ionosphere. J. Geophys. Res. 106, 21305–21336 (2001).
Fox, J. L. & Kliore, A. J. in Venus II: Geology, Geophysics, Atmosphere, and Solar Wind
Environment (eds Bougher, S. W., Hunten, D. M. & Philips, R. J.) 161–188 (Univ.
Arizona Press, Tucson, 1997).
Kliore, A. J. & Luhmann, J. G. Solar cycle effects on the structure of the electron
density profiles in the dayside ionosphere of Venus. J. Geophys. Res. 96,
21281–21289 (1991).
Kliore, A. J., Luhmann, J. G. & Zhang, M. H. G. The effect of the solar cycle on the
maintenance of the nightside ionosphere of Venus. J. Geophys. Res. 96,
11065–11071 (1991).
Taylor, F. W. et al. Structure and meteorology of the middle atmosphere of Venus:
infrared remote sensing from the Pioneer orbiter. J. Geophys. Res. 85, 7963–8006
(1980).
Supplementary Information is linked to the online version of the paper at
www.nature.com/nature.
Acknowledgements We thank H. Svedhem, F. Jansen, the Project Science Team at
ESTEC and the Flight Control Team at ESOC for continuous support. The German
and the US part of VeRa are supported by DLR, Bonn-Oberkassel and by a contract
with NASA, respectively.
Author Information Reprints and permissions information is available at
www.nature.com/reprints. Correspondence and requests for materials should be
addressed to M.P. ([email protected]).
660
©2007 Nature Publishing Group