A warm layer in Venus cryosphere and high-altitude measurements of HF, HCl, H2O and HDO

Full text

Page 1

LETTERS
A warm layer in Venus’ cryosphere and high-altitude
measurements of HF, HCl, H2O and HDO
Jean-Loup Bertaux1,2,5, Ann-Carine Vandaele3, Oleg Korablev4, E. Villard1,2,5, A. Fedorova4, D. Fussen3,
E. Que ´merais1,2,5, D. Belyaev4, A. Mahieux3, F. Montmessin1,2,5, C. Muller3, E. Neefs3, D. Nevejans3, V. Wilquet3,
J. P. Dubois1,2,5, A. Hauchecorne1,2,5, A. Stepanov4,6, I. Vinogradov4, A. Rodin4,7& the SPICAV/SOIR team*
Venus has thick clouds of H2SO4aerosol particles extending from
altitudesof40to60km.The60–100kmregion(themesosphere)is
a transition region between the 4day retrograde superrotation at
the top of the thick clouds and the solar–antisolar circulation in
the thermosphere (above 100km), which has upwelling over the
subsolar point and transport to the nightside1,2. The mesosphere
has a light haze of variable optical thickness, with CO, SO2, HCl,
HF, H2O and HDO as the most important minor gaseous consti-
tuents, but the vertical distribution of the haze and molecules is
poorly known because previous descent probes began their mea-
surements at or below 60km. Here we report the detection of an
extensivelayerofwarmairataltitudes90–120kmonthenightside
that we interpret as the result of adiabatic heating during air sub-
sidence. Such a strong temperature inversion was not expected,
because the night side of Venus was otherwise so cold that it was
named the ‘cryosphere’ above 100km. We also measured the
mesospheric distributions of HF, HCl, H2O and HDO. HCl is less
abundant thanreported 40years ago3.HDO/H2Oisenhancedbya
factor of 2.5 with respect to the lower atmosphere, and there is a
general depletion of H2O around 80–90km for which we have no
explanation.
The use of solar (at the terminator: the circle on the planet that
separates the day side from the night side) and stellar occultation
technique (at night), applied for the first time to the atmosphere of
Venus with the SPICAV/SOIR spectrometers on board the Venus
Express spacecraft, allows us to measure the atmospheric transmis-
sion and to derive information about the vertical structure and com-
position of the 60–140km region. In this region, many processes
(transport,chemistry,temperature, aerosolscondensationandevap-
oration) govern the three-dimensional distribution of haze and
chemical species. In the ultraviolet range (110–310nm) of the
SPICAV ultraviolet spectrometer (a copy of the SPICAM ultraviolet
instrument in orbit around Mars4), the most important absorbers
are CO2(the main Venus atmospheric constituent) at l,200nm,
and aerosol particles of the haze layer at all wavelengths. The CO2
local density and temperature profiles are derived by assuming that
the atmosphere is in hydrostatic equilibrium5(Supplementary
Information).
OnFig.1arerepresentedthetemperatureprofilesobtainedduring
six stellar occultations performed on the night side. They are com-
pared to previous and scarce measurements6–9. All our new profiles
show a large temperature excess (30–70K) with respect to previous
measurements, peaked around 100km. The altitude range 100–
150km has largely been unexplored up to now—the upper limit
for infrared soundings and radio occultation is ,100km, descent
probes have so far measured below this range, and drag measure-
mentsaremadeabove150km.Occultationmeasurementscanprobe
this region efficiently, allowing the discovery of this hot atmospheric
layer. Orbits 95, 96and 98have similar profiles, while orbits 102, 103
and 104 have the same kind of profiles, but with an even higher
maximum temperature. These three orbits are much closer to the
anti-solar point (solar zenith angle, SZA5167–170u).
We interpret this newly found temperature peak to be caused by
adiabatic (or diabatic) heating during air subsidence near the anti-
solarpoint,astheendresultofthesolar–antisolarcirculationpattern
suspected to exist in the thermosphere (100–200km) from the day
side to the night side. Such heating was not well predicted by current
circulation models, although there was evidence of day-to-night
transport and downward vertical transport on the night side: the
emission of NO ultraviolet delta and gamma bands already observed
by Pioneer Venus10,11, and O2emission at 1.27mm discovered from
the ground12(also well detected with VIRTIS/VEX13). These emis-
sions occur when O and N atoms (produced by solar extreme-
ultraviolet photo-dissociation of N2and CO2on the day side at
100–120km) recombine in the night side. However, modelling of
NO ultraviolet emission11,14described N and O atoms as being trans-
ported vertically downwards through the CO2background gas by
eddy diffusion (turbulence).
In such a description, the CO2background gas does not move
vertically, and therefore experiences no adiabatic heating. Our
observations indicate rather that N and O atoms are advected
downwards with the CO2background gas during its descent. Such
a vertical atmospheric motion is an essential ingredient of the
solar–antisolar circulation pattern, which may influence the beha-
viour of the whole mesosphere, because it implies a compensating
upwelling on the day side. The descent velocity may be estimated,
given that at temperature T5165K the night-time infrared cooling
rate is about 100K per day (ref. 2). Assuming a T4dependence, it
would amount to 377K per day for a temperature of 230K as mea-
sured here near the anti-solar point. This cooling rate can be com-
pensated by an equal adiabatic heating rate, corresponding to a
descent velocity V50.43ms21, using the relation dT/dt52CV
where t is time and C is the adiabatic lapse rate (about 10Kkm21;
ref. 15). The negative temperature gradient in the layer at 100–
120km is on average equal to 25Kkm21for orbit profiles 102 to
104 and locally reaches values near the adiabatic lapse rate. This
suggests that the layer at 100–120km is dynamically nearly unstable
and that turbulence may occur, inducing a downward heat flux by
*Lists of participants and affiliations appear at the end of the paper.
1Service d’Ae ´ronomie du CNRS/IPSL, Verrie `res-le-Buisson 91371, France.2Universite ´ Pierre et Marie Curie, 75252, Paris, France.3Belgian Institute for Space Aeronomy, 3 avenue
Circulaire, B-1180 Brussels, Belgium.4Space Research Institute (IKI), 84/32 Profsoyuznaya, 117810 Moscow, Russia.5Institut Pierre Simon Laplace, Universite ´ de Versailles-Saint-
Quentin, 78 Saint-Quentin en Yvelines, 78280 Guyancourt , France.6Faculty of Physics, Moscow State University, GSP-2 119992 Moscow, Russia.7Moscow Institute of Physics and
Technology, Institutsky dr. 141700 Dolgoprudny, Russia.
Vol 450|29 November 2007|doi:10.1038/nature05974
646
Nature
©2007
PublishingGroup

Page 2

mixing of potential temperature and a reinforcement of the peak
temperature.
This is the first time that a temperature inversion is so clearly
identified at this altitude in the night-side atmosphere of Venus.
However, there is a hint of temperature inversion in previous data
at their maximum altitude soundings: a 2–4K increase from 95 to
100km ininfrared spectra8, and a10K increase from90 to 100km in
theradio-occultation profileofMagellan9takenataSZAof109u.We
note that, in our profiles, the higher (the nearer to the anti-solar
point)istheSZA,thehigheristhetemperaturemaximum,thistrend
being confirmed on the low side of the SZA by Magellan (which has
an excellent vertical resolution and sampling). Whether our new hot
profiles are sporadic and patchy, or are permanent and large-scale
features (having escaped detection before) will be clarified by future
Venus Express measurements.
Hydrogen-bearing molecules (HCl, HF, H2O and HDO) are bro-
ken by solar ultraviolet in the altitude range 80–120km, and their
abundanceinthemesosphereisrelevanttotheultimateescaperateof
H, and possible evolution of the Venus atmosphere. HCl and HF
were discovered in 1967 (ref. 3). More recently, the abundance of
HDO (ref. 16) and H2O (ref. 17) above cloud top were found to
experience large time variations (by a factor of 30), even when the
measurements were averaged over the whole disk of Venus, with no
explanationasyet.Almostnothingisknownabouttheverticalprofile
of these species, although it is essential for modelling of H escape. A
new compact high-spectral-resolution infrared instrument, called
Solar Occultation in the InfraRed (SOIR)18was implemented on
Venus Express as an extension of SPICAV, to measure the vertical
distribution of hydrogen-bearing species in the range 60–110km by
the technique of solar occultation at terminator (see the Supple-
mentary Information). SOIR is a new type of spectrometer, with an
echelle spectrometer associated to an acousto-optical tunable filter
for wavelength domain selection, measuring the solar spectrum
and the atmospheric transmittance (Fig. 2) in the infrared region
80
180
200
220
240
80100
100 120 140
SZA (º)
Ref. 7
Pioneer Venus Night Probe 1978
Ref. 6
Orbit 95 (SZA = 121º)
Orbit 96 (SZA = 122º)
Orbit 98 (SZA = 124º)
Orbit 102 (SZA = 167º)
Orbit 103 (SZA = 169º)
Orbit 104 (SZA = 170º)
Radio Occ Magellan (SZA = 109º)
160 180
120140160
Temperature (K)
Temperature (K)
180 200220240
100
Altitude (km)120
140
Figure 1 | Night-side temperature of Venus atmosphere. The mesopause
may be defined at 90km of altitude, separating the mesosphere (,90km)
from the lower thermosphere (.90km). Solid thick curves, obtained by
stellar occultations on the night side from Venus Express (this work), are
compared to previous measurements. The blue squares were obtained
duringthedescentofthePioneerVenusnight-sideprobe.Thethinsolidline
shows sub-millimetre observations from a ground-based radio-telescope,
Venusdiskintegrated6.Thedashedlineshowsmillimetreobservationsfrom
a ground-based radio-telescope, Venus disk integrated7. The thick dotted
linewasobtainedfromaradio-occultationoftheMagellanorbiter9,ataSZA
of 109u. This single measurement shows an increase of temperature at
altitudes above 85km, similar to the ones seen in this work, but less
pronounced. The SPICAV error bars (1s) are indicated for orbit 95 (green)
and are typical of the errors for the other curves. The two groups of
measurements were taken at latitude 39uN (local time ,21:00h,
SZA5121–124u)fororbits95,96,98andlatitude4uS(localtime,23:20h,
SZA5167–170u) for orbits 103, 103 and104.The orbit is24h long.There is
a clear pattern for the value of the maximum temperature (90–100km),
increasing with the SZA, as emphasized in the bottom left plot (the point at
SZA5109u is from Magellan).
2,7052,7102,715
Wavenumber (cm–1)
Transmittance
2,7202,725
70
0
0.2
0.4
0.6
0.8
1
75
80
85
90
HDO
Approximate altitude (km)
Figure 2 | Typicalevolutionofatmosphericspectraltransmittancesthrough
one solar occultation observed by SOIR spectrometer. It is obtained by
determining the ratio of the solar spectrum seen through the Venus
atmosphere to the unattenuated solar spectrum measured above the
atmosphere, at high spectral resolution and a 3 to 6km altitude sampling. At
the beginning of the series, the light path does not cross the atmosphere. No
absorption signatures are present and transmittances are equal to unity. As
the sun sets, the light path goes deeper and deeper into the atmosphere, and
two absorption processes take place: the overall signal decreases owing to
extinction byaerosols,and gaseous absorptionsignaturesappear. Attheend,
thelightpathcrossesthecloudlayerlocatedatanaltitudearound60kmabove
the Venus surface (at 6,051.5km radius) and light is no longer transmitted.
This solar occultation was collected on 26 November 2006 during a sunset.
The selection of a spectral interval is achieved through the acousto-optical
tunablefilter,tunedinthiscasetothe2,703.5–2,727.5cm21spectralrange.In
thisparticularrange, themainabsorptionlines are from HDO (atrio oflines
indicated by arrows), and other features are from weak CO2spectral lines.
NATURE|Vol 450|29 November 2007
LETTERS
647
Nature
©2007
PublishingGroup

Page 3

(2.2–4.3mm) at a resolution of 0.15cm21. At present, using it,
absorption structures of CO2, H2O, HDO, CO, HCL and HF have
been observed with certainty.
Retrieval of the vertical distribution of the various species is done
by comparing the observed transmittance spectra to a full forward
modelsimulation19.Someresultsofthisbest-fitexerciseareshownin
Fig. 3 for HF and HCl. The abundance of HF measured below the
clouds20at 0.00560.002p.p.m. agrees with other measurements21
corresponding to altitudes above 72km (0.006560.0003p.p.m.).
Our SOIR-measured values (0.001 to 0.003p.p.m. at 75–85km) are
significantly lower (by a factor of 3).
Ataltitudesbetween15and30km,lowerthanthoseprobedbythe
SOIR instrument, an HCl mixing ratio of 0.560.15p.p.m. was
obtained in 1989 (ref. 20) while a similar value of 0.660.12p.p.m.
wasfound3in1967abovetheclouds(60km).TheabundancesofHCl
obtained from the SOIR measurements during the occultations of
orbits 136 and 247, shown in Fig. 3a, are slightly different from each
other (by a factor of ,1.5 at 70–75km), and are lower than values
reportedintheliterature(factorof4to6):0.160.03p.p.m.fororbit
136 and 0.1760.03p.p.m. for orbit 247, both at 70–75km. Current
models of photochemistry of the atmosphere of Venus above the
clouds22are assuming a value of 1p.p.m.v. of HCl for this very active
component and should be revised accordingly. Another possible
explanation of our low values is that the solar occultation probed a
high latitude, where thermospheric air depleted in HF and HCl
(because of photo-dissociation at high altitude) may be spiralling
down the polar vortex, along the descending branch of the
Hadley’s cell whose up-leg is at the subsolar point (Y. L. Yung & R.
T. Clancy, personal communication).
Figure 4a shows retrievals of H2O and HDO mixing ratios
obtained simultaneously with the acousto-optical tunable filter
sequence alternating between the H2O spectral window and the
HDO window, and with a window also dedicated to CO2retrieval.
Thethreeprofileswereobtainedover18days,atahighnorthlatitude
(80–75u) terminator. H2O is at 0.3–1p.p.m.v, an intermediate value
withintheverylargerangeof0.1to4p.p.m.vreportedfromprevious
measurements by different techniques16,17at various times. At vari-
ance, our three vertical profiles show little time variability in this
limited sample. There is a marked depletion of H2O in the range
80–90km, for which we have no explanation yet, other than noting
that this altitude range coincides with the mesospheric minimum
temperatureandthetopofthehazelayer.ForH2O,thereisnoabrupt
decreasing at high altitudes (up to 110km), which would be a sign of
local condensation.
The HDO/H2O ratio profiles (Fig. 4b) are quite similar. At the
lower boundary of our measurements (70km), the HDO/H2O ratio
is ,0.1 with an error bar (1s) exceeding 50%, which encompasses
the value of 0.05 measured in the bulk atmosphere at lower alti-
tudes20,23–25(to be compared to the Earth’s value of 1/3,000).
Although we could have expected a decrease of this ratio with
increasing altitude because of preferential condensation of HDO26,
the trend is instead an increase in the HDO/H2O ratio, up to a value
of0.12,about2.5timesthebulkatmospherevalue.Hence,thereisno
effective cold trap preventing the photo-dissociation of HDO in the
upper atmosphere. The observed bulge of HDO at 90–95km, above
the haze and free from extinction of solar flux, might be due to a
lower photo-dissociation rate of HDO (versus H2O), as has been
suggested for Mars from laboratory cross-section measurements27.
Alternatively, it could be a sign that H atoms are escaping to outer
spacefasterthanDatoms(becauseoftheirlargerthermalvelocity):D
atoms left behind in excess will recombine with OH radicals, gen-
erating a downward flow of HDO. If this interpretation is correct, it
would be the first indication of this expected differential escape of H
versus D, acting at present, that could explain the high D/H ratio in
the present atmosphere of Venus. However, while the present 3cm
100
95
90
85
80
Altitude (km)
75
70
65
60
0.0 0.1 0.2 0.3
Vertical mixing ratio (p.p.m.)
HCI
Orbit 136
Orbit 247
Orbit 114
0.4 0.5 0.60.0000 0.0025 0.0050
Vertical mixing ratio (p.p.m.)
HF
a
b
100
0.0075 0.0100
95
90
85
80
75
70
65
60
Figure 3 | HF and HCl mixing ratio vertical profiles retrieved from SOIR
occultations. The mixingratios are computed from the density retrievals of
HF and HCl, divided by the CO2density retrieved from CO2absorption
lines. The number of the orbit analysed is indicated on each graph. a, Two
HCl profiles are compared, taken at orbit 136 (4 September 2006) and 247
(24 December 2006). They are somewhat different in the range 70–75km of
altitude,andthemixingratioseemsto increaseslightlywithaltitudeinboth
profiles. Error bars are 1s. The amount of HCl is much less than is assumed
in the photochemistry models (1p.p.m.), and less than 0.660.12p.p.m.
reported above the clouds3, even after revision of this measurement to
0.4260.07p.p.m. (ref. 28). The vertical bar represents a constant volume
mixing ratio of 0.5p.p.m., as an average of the two derivations3,28from 1967
observations. b, The HF (hydrofluoric acid) volume mixing ratio SOIR
profile at orbit 114 (13 August 2006) is somewhat lower by a factor of ,3
than a previous measurement corresponding to altitudes above 72km
(vertical bar).
70
10−2
75
80
85
90
95
100
105
110
Altitude (km)
Orbit 244
Orbit 251
Orbit 262
Orbit 244
Orbit 251
Orbit 262
75
80
85
90
Altitude (km)
70
95
HDO and H2O
ba
HDO/H2O
10−1
100
0.050.10.150.2 0.2510−2
0
Vertical mixing ratio (p.p.m.)
Ratio
Figure 4 | HDO and H2O mixing ratio, HDO/ H2O vertical profiles. Both
H2O and HDO were measured simultaneously with SOIR during solar
occultationatorbits244,251and262between70and110kmrespectivelyat
polar latitudes 185,83,73u (December 2006 to 8 January 2007). Error bars
are 1s. a, HDO slant densities were deduced using absorption structures
locatedaround2,722.5cm21(Fig.2)andthoseofH2Ofromstructuresinthe
3,832.0–3,852.0cm21spectral interval. After vertical inversion to obtain
local densities, they were divided by the CO2density retrieved
simultaneously from CO2absorption lines to get the volume mixing ratios.
The less abundant curves are for HDO (the three curves on the left). The
curves of both isotopes show little variability, but there is an unexplained
andconsistentH2Odepletionaround85km.b,TheHDO/H2Oratiovertical
profiles are compared for the three orbits. While this ratio is 1/3,000 in
Earth’s sea water, Venus is known to be enriched in HDO by a factor of 150,
as a result of preferential escape of the lighter isotope H20,23–25. The value of
this ratio in the bulk lower atmosphere is measured to be 0.05, while SOIR
finds a higher ratio in the whole range 70–95km. The HDO/H2O ratio
increases with altitude, reaching about 0.12 (a factor of 2.5 above the bulk
ratio). For a tentative explanation of this deuterium-super-enriched layer,
see text.
LETTERS
NATURE|Vol 450|29 November 2007
648
Nature
©2007
PublishingGroup

Page 4

equivalent liquid of water and D/H ratio<0.025 (enrichment 150)
wouldimplyaglobalquantityofwaterofonly4.5mdepthinthepast
comparedtoEarth’s2.8-km-deepocean(iftherewerenoDescape),a
significant escape of D atoms could largely increase this lower limit.
The observed presence of HDO in the photo-dissociation region
indicates that D atoms must be present in the thermosphere, where
they might suffer non-thermal escape mechanisms, as well as H
atoms. It is therefore important to quantify the non-thermal escape
mechanism of H and D atoms.
Received 28 February; accepted 22 May 2007.
1.Dickinson,R.E.&Ridley,E.C.Venusmesosphereandthermospheretemperature
structure: II. Day-night variations. Icarus 30, 163–178 (1977).
Bougher, S. W., Alexander, M. J. & Mayr, H. G. in Venus II (eds Bougher, S. W.,
Hunten, D. M. & Phillips, R. J.) 259–291 (Univ. Arizona Press, Tucson, 1997).
Connes, P., Connes, J., Benedict, W. S. & Kaplan, L. D.Traces of HCl and HF in the
atmosphere of Venus. Astrophys. J. 147, 1230–1237 (1967).
Bertaux, J.-L. et al. SPICAM on Mars Express: Observing modes and overview of
UV spectrometer data and scientific results, J. Geophys. Res. 111, E10S90,
doi:10.1029/2006JE002690 (2006).
Que ´merais, E. et al. Stellar occultations observed by SPICAM on Mars Express.
J. Geophys. Rev. 111, doi:10.1029/2005JE002604 (2006).
Clancy, R. T., Sandor, B. J. & Moriarty-Schieven, G. H. Observational definition of
the Venus mesopause: vertical structure, diurnal variation, and temporal
instability. Icarus 161, 1–16 (2003).
Lellouch, E., Goldstein, J. J., Rosenqvist, J., Bougher, S. W. & Paubert, G. Global
circulation, thermal structure, and carbon monoxide distribution in Venus’s
mesosphere in 1991. Icarus 110, 315–339 (1994).
Zasova,L.V.,Moroz,V.I.,Linkin,V.M.,Khatuntsev,I.V.&Maioriv,B.S.Structure
oftheVenusianatmospherefromsurfaceupto100km.CosmicRes.44,364–383
(2006).
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, 79–94 (1994).
10. Stewart, A. I. F., Ge ´rard, J. C., Rusch, D. W. & Bougher, S. W. Morphology of the
Venus ultraviolet night airglow. J. Geophys. Res. 85, 7861–7870 (1980).
11.Gerard, J. C., Stewart, A. I. F. & Bougher, S. W. The altitude distribution of the
Venus ultra-violet nightglowandimplications onVertical transport. Geophys.Res.
Lett. 8, 633–636 (1981).
12. Connes, P.,Noxon,J.F.,Traub,W. A.&Carleton,N.P.O2(1D)emissionintheday
and night airglow of Venus. Astrophys. J. 233, L29–L32 (1979).
13. Drossart, P. et al. Venus upper atmospheric emissions from VIRTIS spectral
imaging observations. Nature doi:10.1038/nature05xxx (this issue)..
14. Bougher, S. W., Gerard, J. C., Stewart, A. I. F. & Fessen, C. G. The Venus
nitric oxide night airglow—Model calculations based on the Venus
Thermospheric General Circulation Model. J. Geophys. Res. 95, 6271–6284
(1990).
15. Keating,G.M.etal.VIRA(VenusInternationalReferenceAtmosphere)modelsof
Venus neutral upper atmosphere: structure and composition. Adv. Space Res.
5(11), 117–171 (1985).
16. Sandor, B. J. & Clancy, R. T. Water variations in the Venus mesosphere from
microwave spectra. Icarus 177, 129–143 (2005).
17. Gurwell, M. A. Melnick, G. J., Tolls, V., Bergin, E. A. & Patten, B. M. SWAS
observations of water vapor in the Venus mesosphere. Icarus 188, 288–304
(2007).
18. Nevejans, D. et al. Compact high-resolution spaceborne echelle grating
spectrometer with acousto-optical tunable filter based order sorting for the
infrared domain from 2.2 to 4.3 mm. Appl. Opt. 45, 5191–5206 (2006).
19. Vandaele, A. C., Kruglanski, M. & De Mazie `re, M. Simulation and retrieval of
atmospheric spectra using ASIMUT. In ESA Atmospheric Sci. Conf. (Frascati, Italy,
2006) [CD-ROM] 71.1 (ESA SP-628, Noordwijk, 2006).
2.
3.
4.
5.
6.
7.
8.
9.
20. Be ´zard, B.,De Bergh,C.,Crisp, D.&Maillard, J.-P. Thedeep atmosphere ofVenus
revealed by high-resolution nightside spectra. Nature 345, 508–511 (1990).
21. Bjo ¨raker, G.L.etal.Airborneobservations ofthegascomposition ofVenus above
the cloud tops: measurements of H2O, HDO, HF and the D/H and18O/16O
isotopic ratios. Bull. Am. Astron. Soc. 24, 995 (1992).
22. Yung, Y. L. & DeMore, W. B. Photochemistry of the stratosphere of Venus:
Implications for atmospheric evolution. Icarus 51, 199–247 (1982).
23. Donahue, T. M., Hoffman, J. H., Hodges, R. R. & Watson, A. J. Venus was wet—
a measurement of the ratio of deuterium to hydrogen. Science 216, 630–633
(1982).
24. Donahue, T. M., Grinspoon, D. H., Hartle, R. E. & Hodges, R. R. in Venus II (eds
Bougher, S. W., Hunten, D. M. & Phillips, R. J.) 385–414 (Univ. Arizona Press,
Tucson, 1997).
25. De Bergh, C. et al. Deuterium on Venus: observations from Earth. Science 251,
547–549 (1991).
26. Bertaux, J. L. & Montmessin, F. Isotopic fractionation through water vapor
condensation: the deuteropause, a cold trap for deuterium in the atmosphere of
Mars. J. Geophys. Res. Planets 106, 32879–32884 (2001).
27. Cheng, B.-M. et al. Photo-induced fractionation of water isotopomers in the
martian atmosphere. Geophys. Res. Lett. 26, 3657–3660 (1999).
28. Young, L. D. G. High resolution spectra of Venus-A review. Icarus 17, 632–658
(1972).
Supplementary Information is linked to the online version of the paper at
www.nature.com/nature.
Acknowledgements Venus Express is a space mission from the European Space
Agency (ESA). We wish to thank all ESA members who participated in this
successful mission, and in particular H. Svedhem, D. McCoy, O. Witasse,
A.AccomazzoandJ.Louet.WealsothankAstriumforthedesignandconstruction
of the spacecraft, and in particular A. Clochet, responsible for the payload. We
thankD.HinsonforcommunicationoftheMagellanradio-occultationatmospheric
profile, and Y. Yung and T. Clancy for discussions. We thank our collaborators at
Service d’Ae ´ronomie/France, BIRA/Belgium and IKI/Moscow for the design and
fabrication of the instrument. We thank CNRS and CNES for financing SPICAV/
SOIR in France, the Belgian government, Roskosmos and the Russian Academy of
Sciences. The Russian team acknowledges support from the Russian Foundation
for Basic Research, and from the Russian Science Support Foundation.
Author Information Reprints and permissions information is available at
www.nature.com/reprints. The authors declare no competing financial interests.
Correspondence and requests for materials should be addressed to J.-L.B.
(bertaux@aerov.jussieu.fr).
*The SPICAV/SOIR team Jean-Loup Bertaux1, D. Nevejans2, Oleg Korablev3,
F. Montmessin1, Ann-Carine Vandaele2, A. Fedorova3, M. Cabane1, E. Chassefie `re1,
J. Y. Chaufray1, E. Dimarellis1, J. P. Dubois1, A. Hauchecorne1, F. Leblanc1, F. Lefe `vre1,
P. Rannou1, E. Que ´merais1, E. Villard1, D. Fussen2, C. Muller2, E. Neefs2,
E. Van Ransbeeck2, V. Wilquet2, A. Rodin3, A. Stepanov3, I. Vinogradov3, L. Zasova3,
F. Forget4, S. Lebonnois4, D. Titov5, S. Rafkin6, G. Durry7, J. C. Ge ´rard8& B. Sandel9
1Service d’Ae ´ronomie du CNRS/IPSL, Verrie `res-le-Buisson 91371, France.2Belgian
Institute for Space Aeronomy, 3 avenue Circulaire, B-1180 Brussels, Belgium.3Space
Research Institute (IKI), 84/32 Profsoyuznaya, 117810 Moscow, Russia.4Laboratoire de
Me ´te ´orologie Dynamique, 4 place Jussieu, 75252 Paris cedex 05, France.
5Max-Planck-Institut fu ¨r Sonnensystemforschung Max-Planck-Strasse 2, D-37191
Katlenburg-Lindau, Germany.6Southwest Research Institute, Department of
Geophysics, Astrophysics and Planetary Science, 1050 Walnut Avenue, Suite 400,
Boulder, Colorado 80302-5143, USA.7Groupe de Spectrome ´trie Mole ´culaire et
Atmosphe ´rique, Universite ´ de Reims, Champagne-Ardennes B.P. 1039, 51687 Reims
cedex,France.8Universite ´deLie `ge,Institutd’AstrophysiqueetGeophysique—B5c,Alle ´e
du 6 Aout, 17 Sart Tilman, B-4000 Liege, Belgium.9Lunar and Planetary Laboratory,
University of Arizona, 1541 E. University Boulevard, Tucson, Arizona 85721, USA.
NATURE|Vol 450|29 November 2007
LETTERS
649
Nature
©2007
PublishingGroup