Wright State University
CORE Scholar
Physics Faculty Publications
Physics
1998
Solar Cycle Variability of Hot Oxygen Atoms at
Mars
Jhoon Kim
Andrew F. Nagy
Jane L. Fox
Wright State University - Main Campus, [email protected]
Thomas E. Cravens
Follow this and additional works at: http://corescholar.libraries.wright.edu/physics
Part of the Astrophysics and Astronomy Commons, and the Physics Commons
Repository Citation
Kim, J., Nagy, A. F., Fox, J. L., & Cravens, T. E. (1998). Solar Cycle Variability of Hot Oxygen Atoms at Mars. Journal of Geophysical
Research: Space Physics, 103 (A12), 29339-29342.
http://corescholar.libraries.wright.edu/physics/439
This Article is brought to you for free and open access by the Physics at CORE Scholar. It has been accepted for inclusion in Physics Faculty
Publications by an authorized administrator of CORE Scholar. For more information, please contact [email protected].
JOURNAL OF GEOPHYSICALRESEARCH,VOL. 103,NO. A12, PAGES29,339-29,342,
DECEMBER 1, 1998
Solar cycle variability of hot oxygenatoms at Mars
Jhoon
Kim,1Andrew
F. Nagy,
2JaneL. Fox,3 andThomas
E. Cravens
4
Abstract.Thepopulation
of hotoxygen
atomsin theMartianexosphere
isreexamined
using
newlycalculated
hotO production
ratesforbothlowandhighsolarcycleconditions.
Thehot
oxygen
production
rates
areassumed
toresult
fromthedissociative
recombination
ofO• ions.
Thesecalculations
takeintoaccount
thecalculated
vibrational
distribution
of O• andthenew
measured
branching
ratios.Furthermore,
these
calculations
alsoconsider
thevariation
of thedissociative
recombination
crosssection
withtherelativespeedof theparticipating
ionsandelec-
trons,therotational
energy
of the O• ions,andthespread
of theionandelectron
velocities.
These
production
rates
werenextused
inatwo-stream
model
toobtain
theenergy
dependent
fluxof
thehotoxygen
atoms
asa function
ofaltitude.
Finally,thecalculated
fluxattheexobase
wasinputintoanexosphere
model,
based
onLiouville's
theorem,
tocalculate
thehotoxygen
densities
asa functionof altitudein theexosphere
andtheresulting
escape
flux. It wasfoundthathotoxy-
gen
densities
vary
significantly
over
thesolar
cycle;
thecalculated
densities
vary
from
about
2x103
to6x103
cm-3atanaltitude
of 1000km.Theescape
fluxalsovaries
fromabout
3x106
to9x106
cm-2 s-1.
Expressmissionsare creatingnew interestin the upperatmosphere
andexosphere
of Mars. Thispaperdescribes
theresults
of
new
calculations
of
the
hot
oxygen
density
at
Mars
for low
The distribution of nonthermal atmosphericconstituents
and
high
solar
activity
conditions.
andtheirescapeareimportantfor a numberof differentreasons,
suchas atmospheric
evolutionand the interactionof the nonmagneticplanetswith the solarwind. For Venusit hasbeen 2. Model Description
well established
by both observations
[Nagyet al., 1981] and
theoretical calculations[Nagy et al., 1981; McElroy et al.,
Potentialsourcesof hot oxygenat Mars are the dissociative
1982; Nagy and Cravens,1988;Nagyet al., 1990] that hot recombination
of O• andthecharge
exchange
reactions
of O+
oxygenis an important
constituent
in theexosphere.
McElroy ions with H and O. However, it has been demonstratedthat for
1. Introduction
[1972]suggested
thata hotoxygen
population
islikelytobe Venus,
Earth,
andMars,thedominant
source
isdissociative
represent
around
Mars.Since
thisearlysuggestion
a number
of combination
[Nagy
andCravens,
1988;
Nagyetal.,1990;
J.
other
theoretical
model
calculations
[Nagy
andCravens,
1988;Kim,1991'Gerard
etal.,1995;
Richards
etal.,1994;
Hickey
Nagyetal.,1990;
Ip, 1990;
Lainruer
andBauer,
1991]ofhot etal.,1995],
andtherefore
inthisworkit istheonlysource
to
oxygen
andhydrogen
populations
havebeenpublished,
show-beconsidered.
Richards
etal. [1994]andHickey
etal. [1995]
ingthatwhilehotoxygen
is expected
to beanimportant
con- havedrawnattention
to othersources
of thehotO corona
for
stituentin the Martian exosphere,
hot oxygenis not the major the terrestrialcase. Most importantamongthemare quenching
constituent
over any altituderegion. However,no observation andother reactions
of metastable
species,
including
O(1D),
of the hot oxygenpopulationis availableto date. The pres- N(2D)andO+(2D).We havenotincluded
thesesources
herebut
ence of an extendedneutralcoronaplays an importantrole in will include them and dissociativerecombinationof NO+ in a
massloadingand slowingdownthe solarwind at Venusand morecomplete
calculation
to be reported
in thefuture.
Mars.
Recent MHD model calculations have shown that mass
We havecomputed
thenascent
velocitydistribution
of theO
loadingof the solarwind by hot oxygenis importantandis atomsproduced
in dissociative
recombination,
by combining
a
necessaryto explain the observedbow shocklocationsat modelof thevibrational
distribution
of O• withthebranching
Venus and Mars [Bauskeet al., 1998a,b].
ratiosfor the differentpotentialchannelsindicatedin (1), as
The Mars Global Surveyor (MGS), Nozomi, and Mars measured
by Kella et al. [1997]:
O•+e-->O(3p)+ O(3P)+[6.99
eV](0.22)
1Space
Division,KoreanAerospace
Research
Institute,Taejon,
Korea.
-->O (3p)+ O (1D)+[5.02
eV] (0.42)
2Departmentof Atmospheric,
Oceanicand SpaceSciences,
Universityof Michigan,Ann Arbor.
3Institutefor TerrestrialandPlanetaryAtmospheres,
StateUniversity
of New York at StonyBrook,StonyBrook.
-->O (1D)+ O (1D)+[3.06
eV] (0.31)
-•O(3p) + O(1S)+[2.80eV]
(<0.01)
4Department
of Physicsand Astronomy,
Universityof Kansas,
-->O(1D)+ O (1S)+[0.84eV]
(0.05)
Lawrence.
Copyfight
1998bytheAmerican
Geophysical
Union.
Papernumber98JA02727.
0148-0227/98/98JA-02727,$09.00
(1)
wherethe squarebrackets
denotetheexcess
energies
andthe
roundbracketsshowthe branchingratios. We havecarriedout
Monte Carlo calculationsin which the rotationalenergyof the
29,339
29,340
KIM ET AL.: BRIEF REPORT
ion and the initial
velocities
of the ion and electron are chosen
from amonga distributioncharacteristicof the ion and electron
temperatures. Calculationswere carried out for conditionsappropriate to the altitude range in question every 10 km from
130 to 290 km, and the results were interpolated linearly.
Furtherdetailsmay be foundin the work of Fox and Hac [1997].
This calculationis similar to that carried out by Gerard et al.
[1995], who studiedthe effect of the additionalsourcesof hot O '
proposedby Richardset al. [ 1994] andHickey et al. [ 1995] on
the terrestrial hot oxygen corona.
The atmosphericand ionosphericparametersused in the calculationsare shownin Figures l a and lb. The low solar activity model is basedon the neutral density and temperatureprofiles constructedby Fox and Dalgarno [1979] to fit the Viking
1 measurements[Nier and McElroy, 1977]. The ion temperature profile is a smoothedversionof that measuredby the RPA
on Viking 1 [Hanson et al., 1977], and the adoptedelectron
temperaturevaluescomefrom the observations
of Hansonand
Mantas [1988] and the model calculationsof Rohrbaughet al.
[1979]. Further details may be found in the work of Fox
[1993]. The high solar activity neutral temperatureand density
profiles are from the Mars ThermosphericGeneral Circulation
Model (MTGCM) of Bougher and coworkers[Bougheret al.,
1990; Bougher and Shinagawa,1998]. The ion and electron
temperaturesat high solar activity were constructedto be only
slightlylarger than thoseat low solaractivitybut greaterthan
the neutraltemperatureprofile at all altitudes.
An "eroded"ionosphereis the only one assumedfor the low
solar activity case. By erodedionospheres,we mean thosefor
which a lossprocessfor ions is assumedat high altitudes,presumablydue to the interactionof the ionospherewith the solar
wind.
Such an interaction
is characteristic
in situ data, and the plasmapressuremay be large enoughto
withstand
the solar wind.
Therefore
we have constructed both
erodedand nonerodedhigh solar activity models. The low and
high solar activity ionospheres were obtained using the
SC#21REFW and F79050N solar fluxes from Hinteregger
(private communication, 1998 [see also Torr et al., 1979]).
The photoabsorptionand electron impact cross sectionsfor
CO2, O, Ar, N2, CO, 02, andNO usedarethosecompiledby Fox
[1982; 1993]; the H and He crosssectionsare from a compilation of Kirn [ 1991].
The two-streamapproach[Nagy and Banks,1970] wasused
to calculatethe hot oxygenfluxes, as a functionof altitudeand
energy. The crosssectionfor elasticcollisionbetweenhot and
coldoxygenatomswastakento be 1.2x10
-15cm2, assuggestedby McElroy et al. [1982]. The altitude incrementsused
in these calculationsvaried smoothlyfrom 0.2 km at the lower
boundaryof 130 km up to 5 km at 345 km, the upperboundary.
The energygrid usedwas 0.03 eV and coveredthe rangefrom 0
to 6 eV.
The exobase was taken to be at 190 and 210 km for
low and high solar activity cases,respectively. In determining
the exobaselocation we comparedestimatesof the mean free
path and scaleheight,yielding the "classical"value, as well as
evaluating the altitude beyond which the upflowing flux no
longer had a significant effect on the exosphericdensities;
both methodsled to roughly similar estimates. The calculated,
hemispherichot oxygenfluxes at the exobasewere transformed
to an 6nergydistribution,f(E,z):
of bodies without
an intrinsic magnetic field [e.g., Cloutier and Daniell, 1979;
Luhmann,1990]. The erodedionosphere
is modeledby imposing maximum upward velocity boundaryconditionson 11 ions
for which convergenceof the model could be obtained. The
model ionosphereso obtained is one for which the loss rates
are limited by the productionratesof the ions, ratherthan any
specificloss process. Further detailsmay be found in the work
400
soli
of Fox [1997]. The Viking ion density profiles (at low solar
activity) have been found to be reproducibleonly if sucha loss
processis imposed [e.g, Chen et al., 1978; $hinagawa and
Cravens, 1989; Fox, 1993]. At high solar activity, there is no
f(E,z)= {•+(E,z)+•-(E,z)}/v(E)
(2)
where
•+ and•- aretheupward
anddownward,
hemispheric
fluxes, respectively,and v is the oxygen velocity corresponding to energy E. This distributionfunction is then fed into an
exospheremodel, based on Liouville's theorem, which calculates the exosphericdensities.
400
= Smin erod
dotted = Smax,
eroded
__ 02+
_
0 dashed
= Smax,
noneroded
T[
•xxx x•\\
300
TiTe
300
'•200
200
It/
100
!
102
I
I
104
I
I
106
I
I
108
Density
(/cm3)
I
I
1010
,
100
0
,
solid= Stain,
eroded
dashed
=Sma•ellxid
,
,
I
,
1000
,
,
.
I
.
2000
.
,
.
I
,
.
.
3000
.
I
,
.
4000
Temlralare(deg
K)
Figure 1. (a) Adopted atmosphericand ionosphericdensityvaluesfor high and low solar cycle conditions.
(b) Adoptedtemperaturevaluesfor high and low solarcycle conditions.
ß ß
5000
K1M ET AL.: BRIEF REPORT
29,341
3. Results and Discussion
sdki =Snm 190kin, eroded
daled =Smax. 210km, eroded
dashed=Smax. 210km, nonerot•
The calculatedenergy distributionfunctionsat the exobase.
are shown for the three cases considered:low solar activity
erodedand both erodedand nonerodedhigh solar activity cases
in Figure 2. For the sakeof comparisona cold oxygendistribution corresponding
to a temperatureof 195 øK and a densityof
2x106cm-3 anda pseudo
hotcomponent
for anassumed
temperature
of 7000øKanda density
of 104cm-3 arealsoshown.
The visible peaks in the distributionfunction near 2.5 and 3.5
eV correspondto the two branchesof the dissociativerecombination sourcewith the highestbranchingratios.
The calculatedhot oxygendensitiesare plottedas a function
of altitude, for the three different cases,in Figure 3. The hot
oxygen densitiescorrespondingto the low solar activity conditionsare of the samegeneralmagnitudethan the valuescalculated by us earlier [Nagy and Cravens, 1988; Nagy and Kirn,
1990; Zhang et al., 1993] and by Larnrnerand Bauer [1991].
There is a clear and significantincreasein the calculateddensities for the high as comparedto the low solar activity case.
The hot oxygen density estimatesby Kotova et al. [1997],
from the observed solar wind deceleration measurementsby
Phobos2, are nearly an order of magnitudelargerthan even our
high solar cycle values. The calculatedhot oxygen densities
are smaller than the estimatedthermal hydrogendensitiesfor
both low and high solar activity cases. Nevertheless,it has
been shown that the hot oxygen corona plays the dominant
role in massloadingthe solar wind at Mars [Bauske, 1998b]
even thoughthermal hydrogenis the major neutral constituent
in the exosphere. For example,it was shownthat in order to
obtain bow shock positions consistentwith the latest MGS
observations[Acuna et al., 1998] it was necessaryto include
the hot oxygen mass loading process. A comparisonof the
calculatedhot oxygen densitieswith earlier density estimates
for Venus indicatesthat at the higher altitudesthe densitiesat
Mars exceed quite significantly those at Venus. This clearly
demonstratesthat the hot oxygen corona is more extensiveat
Mars, mostlydue to its smallergravity.
[•
daled-Smax.210km,
eroded
105M'•
da•d=Sn•210kn•
n•
'l
lO0O
0
102
103
104
105
Hot OxygenDensity(cm-3)
Figure 3. Calculatedhot oxygen density profiles for high
and low solar cycle conditions.
Integratingthe upwardflux of hot oxygen,with energiesin
excessof the escapeenergy, we evaluatedthe escapeflux per
unit area at the exobase; the results are shown in Table 1. It can
be seenthat the escaperate for the high solar activity caseis
greaterthan that for the low activity one by over a factor of 2.
If one assumesthat the escapeflux is uniform over the entire
exobasesurfacethe total escaperate from the planet is approx-
imately5.3x1024
and 1.3x1025
atomss-1 for thelow andhigh
solar activity cases,respectively. Although the escapeflux is
certainly not homogeneousover the planet, this assumption
does give a useful overall estimate. This escaperate can be
compared with the estimated value for oxygen ion escape,
based on the measurementsobtained by either the ASPERA
[Lundin et al., 1989] or PWS [Nairn et al., 1991] instruments
carriedaboardthe Phobosspacecraft,which was estimatedto be
of theorderof 1025atomss-•. Thusthecurrent
hotoxygen
and
oxygenion escaperatesappearto be of roughlythe sameorder.
Zhang et al. [1993] and Luhmann [1997], estimatedand discussedhow these escapeprocessesmay have varied over the
last 3 Gyr. It was estimatedthat these escapemechanismsmay
accountfor the escapeof up to 30 m of water over this time period; however, some estimatesof the early, planetwide water
inventory have been put as high as 500 to 1000 m [Carr,
1986].
The upcomingNozomi and Mars Expressmissionsto Mars
• 104
___,103
I
'
•..
102
:
Table 1. CalculatedEscapeFlux Values
-,
Low SolarActivity
(Zc=190km)
High SolarActivity
(Zc=210km)
ErodedIonosphere
Ionosphere Ionosphere
Eroded
101
0
1
2
3
Noneroded
Energy (eV)
Escapeflux per
Figure 2. Calculatedenergy distributionof atomic oxygen at unit area,
the exobasefor high and low solar cycle conditionsas indi- atomscm'2 s-1
cated in the figure legend. The smoothlong dash-shortdash
lines indicate a double Maxwellian
fit to the calculated mean
distribution, for the sake of comparison,as describedin the
text.
Total escapeflux,
atomss-1
3.3x106
8.1x106
9.4x106
5.3x1024
1.3x1025
1.5x1025
29,342
KIM ET AL.: BRIEFREPORT
and L. H.
are expectedto be able to addressthe issue of the hot atom Kella, D., P. J. Johnson,H. B. Pedersen,L. Vejby-Christensen,
Andersen, The source of green light emission determinedfrom a
corona and its influence on solar wind interaction processes.
heavy-ionstoragering experiment,Science,276, 1530, 1997.
As quantitativeinformationon hot oxygendensitiesbecomes Kim, J., Model studiesof the ionosphereof Venus:Ion composition,enavailable, it will becomenecessaryto constructmore sophistiergeticsand dynamics, Ph.D. thesis, Univ. of Mich., Ann Arbor,
cated, three-dimensional models of the corona around Mars, in
1991.
order to understandbetterthe variousprocesses,
causingspatial Kim, Y. H., The Jovian ionosphere,Ph.D. thesis, State Univ. of New
York at StonyBrook, 1991.
and temporalvariabilities. Until sucha time the simplemodel Kotova, G., et al., Study of the solarwind decelerationupstreamof the
used in the calculationspresentedin this paper is sufficientto
Martian terminatorbow shock,J. Geophys.Res.,102, 2165, 1997.
demonstratethe importanceof the hot oxygenpopulation.
Lainruer,H., and S. J. Bauer,Nonthermalatmospheric
escapefrom Mars
and Titan, J. Geophys.Res., 96, 1819, 1991.
Acknowledgments. AndrewF. Nagy and ThomasE. Cravensacknowl- Luhmann,J. G., The solarwind interactionwith unmagnetizedplanets:A
tutorial, Geophysical Monograph, edited by pp. 401, AGU,
edge the supportof NASA grantsNAG5-4912 and NAGW-1588 for
Washington,D.C., 1990.
their work, respectively. JaneL. Fox was supportedby NASA grants
NAGW-56007 to the StateUniversityof New York at StonyBrook and Luhmann,J. G., Correctionto "The ancientoxygenexosphereof Mars:
Implicationsfor atmosphereevolution,by Zhang et al., [1993], J.
NAGW-5229 to Wright StateUniversity.
Geophys.Res.,102, 1637, 1997.
The Editor thanksS. Bougherand anotherrefereefor their assistance
Lundin, R., A. Zakharov, R. Pellinen, H. Borg, B. Hultquist, N.
in evaluatingthis paper.
Pissarenko, E. M. Dubinin, S. W. Barabash, I. Liede, and H.
Koskinen,First measurements
of the ionospheric
plasmaescapefrom
References
Mars, Nature, 341, 609, 1989.
McElroy, M. B., Mars: An evolving atmosphere,Science,215, 1614,
1972.
Acuna,M. H., et. al., Magneticfield and plasmaobservations
at Mars:
Initial results of the Mars Global Surveyor mission,Science, 279,
McElroy,M. B., M. J. Prather,andJ. M. Rodriguez,Lossof oxygenfrom
Venus,Geophys.Res.Lett., 9, 649, 1982.
1676, 1998.
Nagy, A. F., and P.M. Banks,Photoelectron
fluxesin the ionosphere,J.
Bauske,R., A. F. Nagy, T. I. Gombosi,D. L. DeZeeuw, K. G. Powell,
Geophys.Res.,75, 6260, 1970.
and J. G. Luhmann,A three-dimensional
MHD studyof solar wind Nagy, A. F., and T. E. Cravens,Hot oxygenatomsin the upper atmomassloading
processes
at Venus:Effectsof photoionization,
electron
spheresof Venusand Mars, J. Geophys.Res.,15, 433, 1988.
impact ionization,and chargeexchange,J. Geophys.Res., 103, in Nagy, A. F., T. E. Cravens,J. H. Yee, and A. I. F. Stewart,Hot oxygen
press, 1998a.
atomsin the upperatmosphere
of Venus,Geophys.
Res.Lett., 8, 629,
Bauske, R., et al., 3D multiscalemassloaded MHD simulationsof the
1981.
solar wind interactionwith VenusandMars, paperpresentedat the Nagy, A. F., J. Kim, and T. E. Cravens,Hot hydrogenand oxygenatoms
32ndCOSPAR ScientificAssembly,Nagoya,1998b.
in the upperatmosphereof Venus and Mars, Ann. Geophys.,8, 251,
Bougher,S. W., andH. Shinagawa,The Mars thermosphere-exosphere; 1990.
Predictionsfor the arrival of Planet-B, Earth Planet. Space,50, 247, Nairn, C. M. C., R. Grard, A. Skalsky,and J. G. Trotignon,Plasmaand
1998.
wave observationsin the night sectorof Mars, J. Geophys.Res., 96,
Bougher,S. W., R. G. Roble,E. C. Ridley,andR. E. Dickinson,The Mars
11227, 1991.
thermosphere,2, General circulationwith coupleddynamicsand Nier, A. O., and M. B. McElroy, Compositionand structureof Mars' upcomposition,
J. Geophys.Res.,95, 14811, 1990.
peratmosphere:
Resultsfrom the massspectrometers
on Viking 1 and
Carr, M. H., Mars: A water-richplanet?,Icarus,56, 187, 1986.
2, J. Geophys.Res.,82, 4341, 1977.
Chen, R. H., T. E. Cravens,and A. F. Nagy, The Martian ionospherein Richards,P. G., M.P. Hickey, and D. G. Torr, New sourcesfor the hot
light of the Viking observations,
J. Geophys.
Res.,83, 3871, 1978.
oxygengeocorona,Geophys.Res.Lett.,21, 657, 1994.
Cloutier,P. A., and R. E. Daniell, An electrodynamical
modelof the solar Rohrbaugh,R. P., J. S. Nisbet,E. Bleuler,andJ. R. Herman,The effect
wind interactionwith the ionospheresof Mars and Venus, Planet.
of energetically
produced
02+ ontheiontemperatures
of theMartian
SpaceSci., 27, 1111, 1979.
thermosphere,
J. Geophys.Res.,84, 3327, 1979.
Fox, J. L., The chemistryof metastablespeciesin the Venusianiono- Shinagawa,H., and T. E. Cravens,A one-dimensional
multispeciesmagsphere,Icarus, 51, 248, 1982.
netohydrodynamicmodel of the dayside ionosphereMars, J.
Fox, J. L., The productionand escapeof nitrogenatomson Mars, J.
Geophys.Res.,94, 6506, 1989.
Geophys.Res.,98, 3297, 1993.
Torr, M. R., D. G. Torr, R. A. Ong, and H. E. Hinteregger,Ionization
Fox, J. L., Upperlimits to the outflowof ionsat Mars: Implicationsfor
frequenciesfor major thermosphericconstituents
as a functionof
atmospheric
evolution,Geophys.
Res.Lett.,24, 2901, 1997.
solarcycle21, Geophys.Res.Lett., 6, 771, 1979.
Fox, J. L., andA. Dalgarno,Ionization,luminosity,andheatingof theup- Zhang,M. H. G., J. G. Luhmann,S. W. Bougherand A. F. Nagy, The
per atmosphere
of Mars,J. Geophys.
Res.,84, 7315, 1979.
ancient oxygen exosphereof Mars: Implicationsfor atmosphere
Fox, J. L., and A. Hac, Spectrumof hot O at the exobases
of the terresevolution,J. Geophys.Res.,98, 10915, 1993.
trial planets,J. Geophys.
Res.,102, 24005, 1997.
Gerard,J. C., P. G. Richards,V. I. Shematovich,and D. V. Bisikalo,The
T. E. Cravens,Departmentof PhysicsandAstronomy,Universityof
importanceof new chemicalsourcesfor the hot oxygengeocorona,
Kansas,Lawrence, KS 66045.
Geophys.Res.Lett., 22, 279, 1995.
J. L. Fox, Institutefor Terrestrialand PlanetaryAtmospheres,State
Hanson,W. B., and G. P. Mantas, Viking electrontemperaturemeasurements:Evidencefor a magneticfield in the Martian atmosphere, Universityof New York at StonyBrook,StonyBrook,NY 11794.
J. Geophys.Res.,93, 7538, 1988.
J. Kim, SpaceDivision, KoreanAerospaceResearchInstitute,52
•anson,W. B., S. Sanatani,
andD. R. Zuccaro,
Themartianionosphere Euon-dong,Yusong-gu,Taejon, Korea.
as observedby the Viking retardingpotentialanalyzer,J. Geophys.
A. F. Nagy, Department of Atmospheric,Oceanic and Space
Res., 82, 4351, 1977.
Sciences, University of Michigan, Ann Arbor, MI 48109.
Hickey,M.P., P. G. Richards,andD. G. Torr, New sources
for the hot ([email protected])
oxygengeocorona:Solar cycle, seasonal,latitudinaland diurnal
variations,J. Geophys.Res.,100, 17377, 1995.
Ip, W. H., The fastatomicoxygencoronaextentof Mars,Geophys.
Res. (ReceivedJuly7, 1998;revisedAugust3, 1998;
Lett., 17, 2289, 1990.
acceptedAugust 17, 1998.)