Mars Research Papers

Ž .Physics of the Earth and Planetary Interiors 117 2000 437–447www.elsevier.comrlocaterpepi

The change of eruption styles of Martian volcanoes and estimatesof the water content of the Martian mantle

Tomonori Kusanagi ), Takafumi MatsuiDepartment of Earth and Planetary Physics, UniÕersity of Tokyo, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan

Received 3 December 1998; accepted 10 May 1999

Abstract

Estimated water contents in the Martian mantle range from 36 ppm to more than 1%. These values are based on thechemical analyses such as hydrous minerals in SNC meteorites and formation models of Mars. This study evaluates thewater content of the Martian mantle using the change with time of volcanic eruption style on Mars as an observationalconstraint. Styles of volcanic activity depend on the volatile content of the magma and the atmospheric pressure. Because alow atmospheric pressure leads to a more explosive volcanic eruption, it has been believed that the volcanism on the currentMartian environment would be very explosive. Our calculations, however, show that, under the current Martian atmosphericconditions, erupted magma cannot entrain the ambient air effectively, so the decrease in temperature of the magma duringascent is small. Consequently, the erupted magma may form a lava-like deposit when it falls back on the ground. Thiseffusive-like style of eruption is a counterpart of clastogenic lava on Mars. On the other hand, numerical calculations under athick CO atmosphere, which may correspond to an ancient Martian atmosphere, reveal a rather explosive eruption style.2

Geological features of earlier stages of Martian history in the Noachian and Hesperian eras suggest that the volcaniceruptions on Mars were explosive then. Effusive eruptions, however, became dominant in more recent times. It has beenwidely accepted that Mars experienced a major climate change. In addition, the release factor of volatiles on Mars has beensuggested to be as small as 0.017–0.112. This may imply that the volatile content has been almost constant throughoutMartian history. Consequently, we assume that this change in eruption style was caused by the change in atmosphericpressure. For a given water content of magma, a major climatic change may lead to a transition in eruption style. If we knowthe atmospheric pressure at the time of this transition, we can calculate the possible range of the volatile content of themantle using our numerical simulations. If the atmospheric pressure on Mars around late Hesperian era is about 1 bar, theestimated values for a typical Martian magma are 0.05–0.25 wt.%, which is within the range of the water content of typicalterrestrial basaltic magmas. q 2000 Elsevier Science B.V. All rights reserved.

Keywords: Eruption styles; Martian volcanoes; Martian mantle

) Corresponding author. e-mail: [email protected]

0031-9201r00r$ - see front matter q 2000 Elsevier Science B.V. All rights reserved.Ž .PII: S0031-9201 99 00112-0

( )T. Kusanagi, T. MatsuirPhysics of the Earth and Planetary Interiors 117 2000 437–447438

1. Introduction

A number of observations from both the groundand space probes such as Vikings 1 and 2, the recentMars Pathfinder, and Mars Global Surveyor haveshown that there are various volcanic features onMars. Some of these volcanoes are much larger thanthe terrestrial counterparts. Particularly, OlympusMons, one of the largest volcanoes on Mars, has aheight of ;27 km and a diameter of ;600 km.Most volcanic features on Mars seem to be theresults of effusive volcanic activities involving lavaflows. Several large volcanoes are similar to shieldvolcanoes on Earth and thought to be composed

Žlargely of low-viscosity basaltic lava e.g., Catter-.mole, 1989; Mouginis-Mark and Wilson, 1992 . Sev-

eral other volcanoes, however, show the signs ofŽexplosive eruptions e.g., Mouginis-Mark et al., 1982,

.1988; Greeley and Crown, 1990 . For example, agroup of Martian volcanoes called ‘‘highland pat-erae’’ have highly eroded slopes suggesting that theirmain bodies may have been constructed of ash or

Žpyroclastic deposits Greeley and Spudis, 1978;Greeley and Crown, 1990; Crown and Greeley,

.1993 .Based on the crater density on the flanks of

Ž .Martian volcanoes, Plescia and Saunders 1979 ar-gued that there were various styles of volcanic activi-ties on the early stage of Martian volcanism but thatonly effusive volcanism survived until the later

Ž .stages. Tanaka 1986 analyzed the stratigraphy ofŽ .Mars in detail and concluded that i the activities of

highland paterae started in the late Noachian epoch,Ž .ii volcanism prevailed over all the volcanic regions

Ž .on Mars during the Hesperian epoch, and iii theactive region was confined to a few locations such asthe Tharsis region from the late Hesperian to theAmazonian epoch. A general trend found here is thatolder volcanoes such as highland paterae may havebeen formed by explosive activity, and more recentvolcanoes are composed of lava flows due to effu-sive activity.

2. The condition for an explosive eruption

An explosive volcanic eruption occurs whenmagma disrupts. The disruption condition is reached

when the volume fraction of gases in the magmaexceeds a critical value. The volume of the exsolvedgases is controlled by the solubilities of the gases tothe magma, which is a function of pressure. Conse-quently, the explosivity of the magma depends onthe gas mass fraction of the magma, the solubility ofeach volatile, and the surface pressure of the planet.The solubility of water into basaltic magma, n , isd

given as a function of pressure P in Pa, as followsŽBurnham, 1975; Wilson and Head, 1981; Stolper

.and Holloway, 1988; Pan et al., 1991 :

n P s6.8=10y8 P 0.7 1Ž . Ž .d

Fig. 1 shows the above relation. Under the surfacepressure of Earth, at least ;0.1 wt.% of water mustbe dissolved into the magma for an explosive erup-tion to be possible. On Mars, however, only about 7ppm of water is enough to cause explosive volcanic

Žactivity because of the low surface pressure Wilson.and Head, 1981 . This means that effusive eruptions

are very hard to produce on Mars because the small-est estimate for the water content of Martian magma

Ž .is 36 ppm Dreibus and Wanke, 1987 . The observed¨morphology of the Martian surface shows that effu-sive volcanism is rather dominant in more recentages, which is not consistent with the considerationabove.

Ž .Wilson and Head 1994 suggested the possibilitythat Hawaiian-type eruptions should have existed onMars. In such eruptions, optically dense fire foun-tains and ineffective entrainment of ambient air keepthe inner parts of fountains hot. Consequently, land-ing magma clots coalesce to form rootless lavas ormagma ponds, even if the magma experienced dis-ruption. The observed large-scale volcanic featureson Mars may have been formed through such erup-tions. To evaluate this possibility and derive theconditions to result in such volcanic activity, wecarried out numerical calculations of volcanic erup-tion processes under Martian conditions.

3. Model

We adopted the model by Sugita and MatsuiŽ .1998 for Martian conditions. This model consistsof two parts, magma rise through the conduit andascent of an eruption cloud in the atmosphere. In

( )T. Kusanagi, T. MatsuirPhysics of the Earth and Planetary Interiors 117 2000 437–447 439

Fig. 1. Conditions for explosive and effusive eruptions. The explosivity of magma depends on the pressure at the planetary surface. Underthe current Martian atmosphere, more than 7 ppm of H O is needed for basaltic magma to erupt explosively. This value is smaller than the2

Ž .least estimate for the H O content of Martian mantle by Dreibus and Wanke 1987 .¨2

both regions, one-dimensional homogeneous steadyflow is assumed. Our model also considers the effect

Ž .of gas bubbles on magma viscosity Jaupart, 1996 .

3.1. The conduit

The generally accepted view of the volcanic erup-tion process is as follows. First, magma in the magmareservoir starts to rise because of buoyancy forcesŽ .Wilson and Head, 1981 . Thus, as it approaches thesurface, the pressure decreases and the volatiles dis-solved in the magma exsolve as gas bubbles, so that

Žthe density of the magma decreases Wilson and.Head, 1981 . As a result, the magma gains buoy-

ancy. The ascent velocity of the magma mainlydepends on both this buoyancy and the wall frictionŽ .McGetchin and Ulrich, 1973 . This is not strictlytrue, as one may include some overpressure due to

Ž .elastic effects for example, around a storage regionand viscous stresses are dominant except at the verylate stages of ascent when the magma is fragmented.However, since it its difficult to take into account the

effect of overpressure, we simply assume that thepressure in the conduit is in equilibrium with that ofthe surroundings. The effect of viscous stresses mightbe taken into account partly in our model because inthe assumption of one-dimensional flow the viscousstress due to horizontal velocity gradient is implicitlyincluded in wall friction.

When the gas volume fraction reaches a criticalvalue, bubbles in the magma come into contact witheach other. Then the magma disrupts and the expan-sion of the gas phase causes an explosive and violenteruption. If magma disruption does not happen, theeruption is effusive and forms a lava deposit. Thecritical gas volume fraction for magma disruption isestimated to be around 70% based on the measure-

Ž .ment of erupted materials on Earth Sparks, 1978 .Ž .Following Sugita and Matsui 1998 , we use the

Ž .equations by Wilson and Head 1981 to describe thebehavior of magma in the conduit. The equations ofmass and momentum conservation are numericallysolved to obtain the vent diameter and the velocity ofmagma. Each equation is given as follows.


Ž .a Mass conservation

Fsp a2rusF sconstant 2Ž .0

where a is the vent radius, r is the bulk density ofmagma, u is the vertical velocity, F is the mass fluxand F is the mass flux at the depth where gas0

Žexsolution initiates hereafter, we call this the exsolu-.tion depth . We assume no mass transfer between

magma and conduit wall. We begin our numericalsimulation by giving the velocity of magma at theexsolution depth, which is assumed to be the termi-nal velocity of bubble-free magma by Wilson and

Ž .Head 1981 .Ž .b Momentum conservation:

du 1 d P fu2

u sy y yg 3Ž .d z r d z 4a

where z is the depth from the planetary surface, f isthe friction coefficient and g is the gravity accelera-tion. The friction coefficient f is given as a functionof the Reynolds number Re by:

64fs q f 4Ž .0Re

where f is the constant of ;0.01.0

The volcanic gas is assumed to be a perfect gasand the mixture density of gas and pyroclasts isgiven by:

1 1yn n Rue es q 5Ž .

r s P

where n is the mass fraction of exsolved gas, s ise

the density of pyroclasts, R is the gas constant of thevolcanic gas and u is the temperature of magma.

The temperature of magma might be consideredto decrease because of the thermal expansion of thegas phase and heat loss through the conduit wall.However, this is not the case. The heat loss throughthe conduit wall is negligible because the time scaleof magma ascent is much smaller than that of the

Ž .thermal conduction Wilson and Heslop, 1990 . Thetemperature change due to bubble expansion is alsovery small because the mass fraction of gas phase ismuch smaller than that of the pyroclasts and thesurrounding magma acts as an effective heat bufferŽ .Sparks, 1978 . Consequently, the magma rise pro-cess may be assumed to be isothermal.

The motion of magma is controlled by manyparameters, including magma viscosity, temperature,density, vent geometry, the depth of the magmareservoir, the volatile content and its solubility tomagma. We use the values of basaltic magma as themagma viscosity, temperature and density and weconsider water as the volatile. Vent geometry affects

Ž .the motion of magma. Giberti and Wilson 1990studied the influence of geometry on the ascent ofmagma in open fissures. However, we simply as-sumed that the conduit is a circular tube. The radiusof the conduit is considered to change according tothe pressure of the magma, which is assumed to beequal to the lithostatic load in this model.

On the other hand, this assumption may not bevalid under the low Martian surface pressure. Thus,we test the effect using the model by Jaupart and

Ž .Tait 1989 . This model is the opposite extreme ofŽ .the model by Wilson and Head 1981 and assumes

that the conduit wall is rigid and the radius of theconduit is constant from the magma chamber to thevent. The pressure of the magma is usually higherthan the lithostatic pressure. Because of the geometryof the conduit, the mixture of gas and melt cannot beaccelerated beyond a critical value. This is called thechoking velocity. The choking velocity, u , is givenc

by the sound velocity of the mixture of gas phaseŽ .and pyroclasts Jaupart, 1996 :

y1r2d ru s 6Ž .c ž /d P

where P is the magma pressure at the vent, r is themixture density of gas phase and pyroclasts at the

Ž . Ž .vent. Using Eq. 5 , Eq. 6 is expressed by:

n RTP 1 1 ( eu s y n q 7Ž .(c e½ 5' ž /s PnRT ( e

where T is the temperature of magma, n is thee

exsolved gas mass fraction, s is the density of thepyroclasts, and R is the gas constant of the volcanicgas. The relationship between the choking velocityand the exsolved gas mass fraction, n , is shown ine

Fig. 2. This result is almost independent of theatmospheric pressure because it is very small com-pared with the vent pressure in this case. For mostcases, the eruption velocity is less than ;150 mrs


Fig. 2. Relationship between the exsolved gas mass fraction andthe choking velocity. Pressure at the vent is 105 and 107 Pa,respectively.

except when the vent pressure is extremely high andexsolved gas fraction is small.

After magma erupts out of the vent, it undergoessubsequent pressure release through a series ofshocks. This increases the vertical velocity and de-creases magma temperature. Since this last processtends to fill the difference between the eruption

Žconditions deduced from the two treatments e.g.,.velocity , the eruption conditions calculated from the

Ž .model of Jaupart and Tait 1989 are not very differ-ent from those calculated from the model of Wilson

Ž .and Head 1981 except that the eruption tempera-ture is somewhat lower. Thus, we adopt the model

Ž .by Wilson and Head 1981 for the following calcu-lation.

3.2. The eruption cloud

The eruption cloud entrains the ambient air duringascent in the atmosphere. The entrained air is heatedby the hot pyroclasts in the eruption cloud andexpands. If the eruption cloud has enough heat andinitial eruption velocity, the cloud gains buoyancy.Once this occurs, the cloud rises very high and formsa convective eruption column. Otherwise, the cloudcollapses and falls back on the ground surface, form-ing a pyroclastic flow fed by a fountain-like structureover the vent.

This process has been studied theoretically by aŽnumber of researchers e.g., Wilson et al., 1980;

.Sparks, 1986 . In this study, we use the formulationŽ .by Woods 1988 with modification by Sugita and

Ž .Matsui 1998 . Equations of the conservations ofmass, momentum and energy are incorporated to-gether with the temperature dependence of the spe-cific heat of gas and pyroclasts.

The equation of mass conservation is given by:

d Fs2p ar u 8Ž .air ed z

where z is the height from the surface, F is the massflux, a is the radius of the eruption cloud, r is theair

density of ambient atmosphere and u is the inwarde

velocity of the surrounding atmosphere. The massflux F is defined as:

Fsp a2ru 9Ž .where u is the vertical velocity and r is the bulkdensity of the eruption cloud. The density r is givenby:

1 1yn nRus q 10Ž .

r s P

where R is the average gas constant of volcanic gasand entrained air, u is the temperature of the erup-tion cloud, n is the mass fraction of the gas phase inthe eruption cloud. Since the mass flux of pyroclasticmaterial is conserved, the relationship between theinitial gas mass fraction n and the gas mass fraction0

at the height z is given by:

1yn Fs 1yn F 11Ž . Ž . Ž .0 0

or

F0ns1y 1yn 12Ž . Ž .0 F

where F is the mass flux at the vent.0ŽThe equation of motion is given by Woods,

.1988 :2du u r ryrair air

u sy y g 13Ž .(d z 8a r r

The momentum equation can also be expressedbased on the buoyancy force as follows:

duF2syg ryr p a 14Ž . Ž .aird z


Ž . Ž . Ž .By substituting Eqs. 9 and 13 into Eq. 14 , theequation of mass conservation can be expressed byŽ .Sugita and Matsui, 1998 :

d F 1s pr uF 15Ž .( aird z 8

This equation is used as the equation of mass conser-Ž .vation instead of Eq. 8 .

ŽThe energy equation is given by Sugita and.Matsui, 1998 :

d u2F d FH u Fq sH T yFg 16Ž . Ž . Ž .air½ 5d z 2 d z

where T is the temperature of the ambient atmo-sphere. H and H is the enthalpy of the eruptionair

cloud and the ambient atmosphere, respectively.

Ž . Ž .Fig. 3. Heights and temperatures at the top of eruption clouds. a and b are the results for the current Martian conditions. We see a clearŽ .distinction between convective eruption clouds and pyroclastic flows. Compared with the calculated results for Earth c and d , the height of

the fountain feeding a pyroclastic flow on Mars is very small, and the temperature decrease at its top is also small.


Since the entrained ambient air dominates themass of an eruption cloud, the eruption style iscontrolled by the vertical structure of the planetaryatmosphere. Since the change in the specific heatalso plays a very important role in the dynamics of

Ž .the eruption cloud Sugita and Matsui, 1998 , thiseffect is also taken into account.

By using the result of the calculation of themagma flow in the conduit as the initial condition,

Žwe can estimate how erupted material rises i.e.,.height and thermal structure of an eruption cloud .

3.3. Combined calculation

Combination of the calculations of two stagesgives the volcanic eruption conditions on Mars. Theconditions are more complex if we consider a col-lapsed eruption column forming a fountain over the

Ž .vent as discussed by Wilson and Heslop 1990 .However, for simplicity, we do not take into accountsuch effects. In this study, the eruption velocity andexsolved gas fraction are derived from the numericalcalculation of the magma ascent in the conduit inwhich the water content of the magma, the magmatemperature, and the vent radius are given as freeparameters. Using these values, the calculation forthe dynamics of an eruption cloud is carried out to

Žcharacterize its nature i.e., the temperature and den-sity of the eruption cloud at each height and the

.ultimate height which the eruption cloud reaches .Thus, we can evaluate the influence of the watercontent on the eruption style through these calcula-tions.

4. Calculation condition and results

We assume basaltic magma with a temperature of1400 K. Although the viscosity of bubbleless magmadepends greatly on the water content, we assume thatit is 100 Pa s for simplicity. The densities of themagma and the country rock used in this study are2500 and 2700 kg my3, respectively. Under theseassumptions, we carried out calculations for twoatmospheric conditions, the thin current Martian at-mosphere and a thicker CO atmosphere.2

4.1. Eruption under current Martian atmosphere

ŽThe current Martian atmosphere is very thin ;.600 Pa and mainly composed of CO . Because of2

the seasonal variation due to the formation of thepolar caps, the pressure fluctuates by 50%. Occa-sional sand storms also alter the pressure by absorb-ing the sunlight.

The temperature structure of Martian atmosphereis derived from the data obtained by Mariner 9,Vikings 1 and 2, and Mars Pathfinder. We used thetemperature structure under a relatively dust-free

Žcondition observed by the Viking landers Seiff and.Kirk, 1977 . The pressure structure is assumed to be

hydrostatic.The result of the calculation for this temperature

profile is shown in Fig. 3. Fig. 3a shows that there isa critical gas mass fraction for a given vent radius,above which the height of the cloud increasesabruptly to form a convective eruption column.

Fig. 4. Relationship between the water mass fraction of the magma and eruption styles under the current Martian atmosphere.


Compared with the result of a calculation for theequivalent eruption on Earth, the height of the col-lapsed fountain after it gets stable is very low onMars. This is because the Martian atmosphere is sothin that the density of the eruption cloud does notdecrease by entrainment of the ambient air into theeruption cloud. In addition, the temperature decreasebetween the vent and the top of the collapsed columnis small. This is also because the cooling due toentrainment is less effective. Consequently, the erup-tion cloud with low initial velocity stops rising be-fore it gets cooled by entraining the ambient air.Thus, the pyroclastic material in such an eruptioncloud will fall on the ground while its temperature isstill high. If the pyroclasts are still molten, they willform a lava-like flow. This kind of eruption is ob-served in Hawaiian type volcanic activity and lavasmade in such activities are called clastogenic lavasŽ .Francis, 1993 . On Earth, clastogenic lava is formedat a high eruption rate when the dense lava fountainprevents pyroclasts from being cooled by radiationŽ .Wilson and Head, 1994 .

This is a kind of eruption that is caused bymagma disrupted in the conduit but that leaves alava-like deposit. The condition to cause such aneruption is illustrated in Fig. 3b and d. We assumethat the clastogenic lava is formed when the tempera-ture decrease during ascent is less than 50 K. Undera thin atmospheric condition, this kind of eruption ismore probable.

Taking this clastogenic lava into consideration,Ž .the condition to form a lava or lava-like deposit on

Mars is that magma contains -0.25 wt.% water.This value is comparable to the average water con-tent of MORB, about 0.2 wt.%. Fig. 4 shows therelationship between gas mass fraction of magmaand eruption style. The overlapping part displays theuncertainty due to the dependence of the eruptionstyle on the vent size.

4.2. Eruption under 1-bar Martian atmosphere

The southern hemisphere of Mars has uniquefluvial features that are thought to have been formed

Ž .by running water e.g., Baker, 1982 . Continuousflow of water requires a relatively high surface tem-

Ž .perature Pollack et al., 1987 . Thus, Mars may oncehave had a thicker atmosphere than today.

We used a one-dimensional radiative–convectiveŽ .equilibrium model Nakajima et al., 1992 to de-

scribe this thicker atmosphere. The composition isassumed to be 100% CO and the surface pressure is2

1 bar.Fig. 5 shows the result of eruption calculations

under a thick CO atmosphere. Pyroclastic flows2

reach relatively high altitudes and are cooled effec-tively. These flows are unlikely to form lava-likedeposit when they fall on the ground. The eruptionstyle under this thick atmosphere is shown in Fig. 6.

Ž . Ž .Fig. 5. a Heights and b temperatures at the top of eruptionclouds under an ancient 1 bar CO atmosphere.2


Fig. 6. Relationship between the water mass fractions of magma and eruption styles under an ancient 1 bar CO atmosphere. The range of2

water mass fraction of magma that can result in a lava-forming eruption is almost the same as that under the current thin atmosphere.

A comparison of Figs. 4 and 6 suggests that theŽconditions for lava-forming eruptions both gas-free

.eruptions and clastogenic lava-forming eruptionsunder an ancient 1 bar atmosphere are not so differ-ent from those in the current atmosphere despite thefact that the water content to cause magma disrup-tion is about one order of magnitude larger. Theminimum magma water content to generate explo-sive eruptions under a 1-bar atmosphere is about0.05 wt.%.

5. Water content of the Martian mantle

We can evaluate the water content of Martianmantle based on the geological features and thehistorical change in the volcanic eruption styles.There are two major candidates for the cause of suchchanges in volcanism. One is the change in volatilecontent in the Martian mantle. If water in the Mar-tian mantle has decreased with time, it would havereduced the explosivity of the Martian volcanoes.The other is the change in the atmospheric condi-tions. As was shown in the numerical results, thisalters the nature of the eruption condition consider-

Ž .ably. Scambos and Jakosky 1990 estimated that theŽ .release factor of nonradiogenic volatiles e.g., water

from the Martian interior since the end of its forma-tion is 0.017–0.112. Such a release factor is toosmall to cause a change in volatile content in theMartian mantle that is sufficient to change the style

Žof volcanic eruptions by itself a release factor of 0.5.is needed from the calculation above .

This suggests that the change in eruption style isdue to the change in atmospheric conditions onMars. On the basis of this assumption, we can obtaina lower limit for the water content required forexplosive volcanism under a thick ancient Martian

Ž .atmosphere 1 bar pressure , and similarly the upperlimit required for more effusive volcanism under thecurrent thin atmosphere. We have estimated thesevalues in Sections 4.1 and 4.2. The estimated lowerlimit is 0.05 wt.% and the upper limit is 0.25 wt.%.

However, the estimated water content may be oneorder of magnitude smaller when we consider thefact that basalt typically represent 10% partial melt-ing. In this case, the value of 0.05–0.25 wt.% mightbe regarded as a maximum estimate.

In former studies, the water content of the Mar-tian mantle has been evaluated from chemical infor-

Ž .mation. Dreibus and Wanke 1987 estimated the¨water content of the Martian mantle to be 36 ppmfrom two component model for the formation ofMars. This estimated value is far smaller than that of

Žterrestrial basaltic magma 0.2–1 wt.% on averageŽ ..Scarth, 1994 .

An alternative way to estimate the water contentof the Martian mantle is from the SNC meteorites.Simple measurement of the water contents in SNCmeteorites gives a value between 130 and 350 ppmŽ .McSween and Harvey, 1993 . The magmas thatformed the SNC meteorites, however, are thought tohave experienced degassing on their way from themagma reservoir to the Martian surface, so this valuemay not give the proper estimate for the Martian

Ž .mantle. Treiman 1985 estimated that the amphibole


content occupies 5–10% of the inclusions in someSNC meteorites and concluded that the originalmagma must have contained at least 0.1–0.2 wt.%water, because the amphibole contains ;2 wt.%water. On the other hand, by considering the solidifi-cation process of the magma which formed the SNCmeteorites, the water content of Martian mantle was

Žestimated to be about 1.4 wt.% Johnson et al., 1991;.McSween and Harvey, 1993 .

Our estimate for the water content of MartianŽ .mantle 0.05–0.25 wt.% is in the range of the

estimates based on SNC meteorites. This is consis-tent with the presumption that SNC meteorites areMartian igneous rocks ejected by the impacts ofother meteorites. The fact that previous chemicalevaluations for water content and estimates based ongeologic features in this study show remarkableagreement reinforces the validity of the assumptionthat the change in volcanic eruption style is causedby epochal climate change on Mars.

Acknowledgements

We thank S. Sugita for his valuable comments onthe early version of this paper, which is useful forthe improvement of the paper. We appreciate thekind and helpful reviews of Lionel Wilson and ananonymous reviewer.

References

Baker, V.R., 1982. The Channels of Mars. Univ. of Texas Press.Burnham, C.W., 1975. Water and magmas; a mixing model.

Geochim. Cosmochim. Acta 39, 1077–1084.Cattermole, P., 1989. Planetary Volcanism: A Study of Volcanic

Activity in the Solar System. Ellos Horwood, 443 pp.Crown, D.A., Greeley, R., 1993. Volcanic geology of Hadriaca

Patera and eastern Hellas region of Mars. J. Geophys. Res. 98,3431–3451.

Dreibus, G., Wanke, H., 1987. Volatiles on Earth and Mars: a¨comparison. Icarus 71, 225–240.

Francis, P., 1993. Volcanoes: A Planetary Perspective. ClarendonPress.

Giberti, G., Wilson, L., 1990. The influence of geometry on theascent of magma in open fissures. Bull. Volcanol. 52, 515–521.

Greeley, R., Crown, D.A., 1990. Volcanic geology of TyrrhenaPatera, Mars. J. Geophys. Res. 95, 7133–7149.

Greeley, R., Spudis, P.D., 1978. Volcanism in the cratered terrainhemisphere of Mars. Geophys. Res. Lett. 5, 453–455.

Jaupart, C., 1996. Physical models of volcanic eruptions. Chem.Geol. 128, 217–227.

Jaupart, C., Tait, S., 1989. Dynamics of eruptive phenomena. In:Ž .Nicholls, J., Russel, J.K. Eds. , Modern Methods of Igneous

Petrology: Understanding Magmatic Processes. Mineral. Soc.Am., Rev. Mineral., Vol. 24, pp. 213–238.

Johnson, M.C., Rutherford, M.J., Hess, P.C., 1991. Chassignypetrogenesis: melt compositions, intensive parameters, and

Ž .water contents of Martian ? magmas. Geochim. Cosmochim.Acta 55, 349–366.

McGetchin, T.R., Ulrich, W.G., 1973. Xenoliths in maars anddiatremes with influences for the Moon, Mars and Venus. J.Geophys. Res. 78, 1833–1853.

McSween, H.Y. Jr., Harvey, R.P., 1993. Outgassed water onMars: constraints from melt inclusions in SNC meteorites.Science 259, 1890–1892.

Mouginis-Mark, P.J., Wilson, L., 1992. The physical volcanologyof Mars. In: Kieffer, H.H., Jakosky, B.M., Snyder, C.W.,

Ž .Matthews, M.S. Eds. , Mars, pp. 424–452.Mouginis-Mark, P.J., Wilson, L., Head, J.W. III, 1982. Explosive

volcanism on Hecates Tholus, Mars: investigation of eruptionconditions. J. Geophys. Res. 87, 9890–9904.

Mouginis-Mark, P.J., Wilson, L., Zimbelman, J.R., 1988. Poly-genic eruptions on Alba Patera, Mars. Bull. Volcanol. 50,361–379.

Nakajima, S., Hayashi, Y., Abe, Y., 1992. A study on the‘‘runaway greenhouse effect’’ with a one-dimensional radia-tive–convective equilibrium model. J. Atmos. Sci. 49, 2256–2266.

Pan, V., Holloway, J.R., Hervig, R.L., 1991. The pressure andtemperature dependence of carbon dioxide solubility in tholei-itic melts. Geochim. Cosmochim. Acta 45, 1587–1595.

Plescia, J.B., Saunders, R.S., 1979. The chronology of Martianvolcanoes. Proc. Lunar Planet. Sci. Conf. 10th, pp. 2841–2859.

Pollack, J.B., Kasting, J.F., Richardson, S.M., Poloakoff, K.,1987. The case for a wet, warm climate on early Mars. Icarus71, 203–224.

Scambos, T.A., Jakosky, B.M., 1990. An outgassing release factorfor nonradiogenic volatiles on Mars. J. Geophys. Res. 95,14779–14787.

Scarth, A., 1994. Volcanoes. Texas A&M Univ. Press.Seiff, A., Kirk, D.B., 1977. Structure of the atmosphere of Mars

in summer at mid-latitudes. J. Geophys. Res. 82, 4364–4378.Sparks, R.S.J., 1978. The dynamics of bubble formation and

growth in magmas: a review and analysis. J. Volcanol.Geotherm. Res. 3, 1–37.

Sparks, R.S.J., 1986. The dimensions and dynamics of volcaniceruption columns. Bull. Volcanol. 48, 3–15.

Stolper, E., Holloway, J.R., 1988. Experimental determination ofthe solubility of carbon dioxide in molten basalt at lowpressure. Earth Planet. Sci. Lett. 87, 397–408.

Sugita, S., Matsui, T., 1998. Can ash-fall type eruption occur onVenus? Implication to condition of Venus interior. J. Geophys.Res., submitted.

Tanaka, K.L., 1986. The stratigraphy of Mars. Proc. Lunar Planet.Sci. Conf. 17th, pp. E139–E158.

Treiman, A.H., 1985. Amphibole and hercynite spinel in Sher-


gotty and Zagami: magmatic water, depth of crystallization,and metasomatism. Meteoritics 20, 229–243.

Wilson, L., Head, J.W. III, 1981. Ascent and eruption of basalticmagma on the Earth and Moon. J. Geophys. Res. 86, 2971–3001.

Wilson, L., Head, J.W. III, 1994. Mars: review and analysis ofvolcanic eruption theory and relationships to observed land-forms. Rev. Geophys. 32, 221–263.

Wilson, L., Heslop, S.E., 1990. Clast sizes in terrestrial and

Martian ignimbrite lag deposits. J. Geophys. Res. 95, 17309–17314.

Wilson, L., Sparks, R.S.J., Walker, G.P.L., 1980. Explosive vol-canic eruptions: IV. The control of magma properties andconduit geometry on eruption column behavior. Geophys. J. R.Astron. Soc. 63, 117–148.

Woods, A.W., 1988. The fluid dynamics and thermodynamics oferuption columns. Bull. Volcanol. 50, 169–193.

© 1999 Macmillan Magazines Ltd

lavas 5 km thick, then our estimates of the total quantity oflava erupted over the past ,4 Gyr (including volcanic edi®ces)must be revised from the 7 3 107 km3 calculated previously20 to5 3 108 km3. Alternatively, the lava could be less extensive if theformation of Valles Marineris was intimately associated with thepresence of a thick lava sequence, which thins rapidly away from thecanyons. If deep layering is con®ned to the area of a rectangleenclosing the canyons (,4 3 106 km2) and is 10 km thick, then thevolume is 4 3 107 km3. This alone greatly exceeds a previousestimate of 8 3 106 km3 of magma extruded in the Late Noachian20.

We conclude that volcanism on early Mars was probably muchmore voluminous than previously documented, and that it musthave affected the climate and near-surface environment. Pollacket al.21 proposed that a warm, wet climate on early Mars wassustained by a thick CO2 atmosphere, which must be continuouslyresupplied or recycled to balance loss of CO2 to carbonates. Twomechanisms for recycling the CO2 have been proposed: extensivevolcanism21 and impacts22. If the recycling was mainly from impacts,then the warm, wet conditions corresponded to the time (on Earth)of heavy bombardment and the impact frustration of life23.Extensive volcanism on Mars could have maintained a thickatmosphere for a signi®cant period of time after the heavybombardment21. The layers seen by MOC provide evidence forvoluminous volcanism; but a thick atmosphere could have beensustained only if suf®cient carbonates exist in the crust of Mars,which has not yet been con®rmed16. M

Received 16 September; accepted 21 December 1998.

1. Lucchitta, B. K. et al. in Mars (eds Kieffer, H. H., Jakosky, B. M., Snyder, C. M. & Matthews, M. S.)

453±492 (Univ. Arizona Press, Tucson, 1992).

2. Tanaka, K. L., Scott, D. H. & Greeley, R. in Mars (eds Kieffer, H. H., Jakosky, B. M., Snyder, C. M. &

Matthews, M. S.) 345±382 (Univ. Arizona Press, Tucson, 1992).

3. Clifford, S. M. A model for the hydrologic and climatic behavior of water on Mars. J. Geophys. Res. 98,10973±11016 (1993).

4. Tanaka, K. L. & Golombek, M. P. Martian tension fractures and the formation of graben and collapse

features at Valles Marineris. Proc. Lunar Planet. Sci. Conf. 19, 383±396 (1989).

5. Davis, P. A. & Golombek, M. P. Discontinuities in the shallow Martian crust at Lunae, Syria, and Sinai

Plana. J. Geophys. Res. 95, 14231±14248 (1990).6. Zuber, M. T. & Aist, L. L. The shallow structure of the Martian lithosphere in the vicinity of the ridged

plains. J. Geophys. Res. 95, 14215±14230 (1990).

7. Malin, M. C. et al. Mars Observer Camera. J. Geophys. Res. 97, 7699±7718 (1992).

8. Malin, M. C. et al. Early views of the Martian surface from the Mars Orbital Camera of Mars Global

Surveyor. Science 279, 1681±1685 (1998).9. Albee, A. L., Palluconi, F. D. & Arvidson, R. E. Mars Global Surveyor mission: Overview and status.

Science 279, 1671±1672 (1998).

10. Lucchitta, B. K. Morphology of chasma walls, Mars. J. Res. US Geol. Surv. 6, 651±662 (1978).

11. Geissler, P. E., Singer, R. B. & Lucchitta, B. K. Dark materials in Valles Marineris: Indications of the

style of volcanism and magmatism on Mars. J. Geophys. Res. 95, 14399±14413 (1990).12. Scott, D. H. & Tanaka, K. L. Geologic Map of the Western Equatorial Region of Mars, Scale 1:15,000,000

(Misc. Inv. Ser. Map I-1802-A, US Geol. Surv., Denver, 1986).

13. Witbeck, N. E., Tanaka, K. E. & Scott, D. H. Geologic Map of the Valles Marineris Region of Mars, Scale

1:2,000,000 (Inv. Ser. Map I-2010, US Geol. Surv., Denver, 1991).14. Erard, S. et al. Spatial variations in composition of the Valles Marineris and Isidis Planitia regions of

Mars derived from ISM data. Proc. Lunar Planet. Sci. Conf. 21, 437±456 (1991).

15. Self, S., Thordarson, T. & Keszthelyi, L. in Large Igneous Provinces (eds Mahoney, J. J. & Cof®n, M. F.)

381±410 (Am. Geophys. Union, Washington, D. C., 1997).

16. Christensen, P. R. et al. Results from the Mars Global Surveyor thermal Emission Spectrometer.Science 279, 1692±1698 (1998).

17. Schubert, G., Solomon, S. C., Turcotte, D. L., Drake, M. J. & Sleep, N. H. in Mars (eds Kieffer, H. H.,

Jakosky, B. M., Snyder, C. M. & Matthews, M. S.) 147±183 (Univ. Arizona Press, Tucson, 1992).

18. Carr, M. H. Water on Mars (Oxford Univ. Press, New York, 1996).

19. Craddock, R. A., Maxwell, T. A. & Howard, A. D. Crater morphometry and modi®cation in the SinusSabaeus and Margaritifer Sinus regions of Mars. J. Geophys. Res. 102, 13321±13340 (1997).

20. Greeley, r. & Schneid, B. D. Magma generation on Mars: Amounts, rates, and comparisons with Earth,

Moon, and Venus. Science 254, 996±998 (1991).

21. Pollack, J. B., Kasting, J. F., Richardson, S. M. & Poliakoff, K. The case for a wet, warm climate on early

Mars. Icarus 71, 203±224 (1987).22. Carr, M. H. Recharge of the early atmosphere of Mars by impact-induced release of CO2. Icarus 79,

311±327 (1989).

23. Maher, K. A. & Stevenson, D. J. Impact frustration of the origin of life. Nature 331, 612±614 (1988).

24. Topographic Maps of the Polar, Western, and Eastern regions of Mars (Misc. Inv. Ser. Map I-2160, US

Geol. Surv., Denver, 1991).25. Fanale, F. P. Martian volatiles: Their degassing history and geochemical fate. Icarus 28, 179±202

(1976).

26. Soderblom, L. A. & Wenner, D. B. Possible fossil water liquid±ice interfaces in the Martian crust.

Icarus 34, 622±637 (1978).

27. Treiman, A. H., Fuks, K. H. & Murchie, S. Diagenetic layers in the upper walls of Valles Marineris,Mars: Evidence for drastic climate change since the mid-Hesperian. J. Geophys. Res. 100, 26339±26344

(1995).

Acknowledgements. We thank L. Keszthelyi for discussions, and M. T. Zuber and N. G. Barlow forcomments on the manuscript. This work was supported by the MGS project.

Correspondence and requests for materials should be addressed to A.S.M. (e-mail: [email protected]).

letters to nature

586 NATURE | VOL 397 | 18 FEBRUARY 1999 | www.nature.com

Evidence for recent volcanismonMars fromcratercountsWilliam K. Hartmann*, Michael Malin², Alfred McEwen³,Michael Carr§, Larry Soderblomk, Peter Thomas¶,Ed Danielson#, Phillip JamesI & Joseph Veverka¶

* Planetary Science Institute, Tucson, Arizona 85705, USA² Malin Space Science Systems, San Diego, California 92191, USA³ Lunar and Planetary Laboratory, University of Arizona, Tucson, Arizona 85721,

USA

§ US Geological Survey, Menlo Park, California 94025, USA

kUS Geological Survey, Flagstaff, Arizona 86001, USA¶ Cornell University, Ithaca, New York 14853, USA

# California Institute of Technology, Pasadena, California 91125, USAI University of Toledo, Toledo, Ohio 43606, USA

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Impact craters help characterize the age of a planetary surface,because they accumulate with time. They also provide usefulconstraints on the importance of surface erosion, as such pro-cesses will preferentially remove the smaller craters. Earlierstudies of martian crater populations revealed that erosion anddust deposition are important processes on Mars1±6. They dis-agreed, however, on the age of the youngest volcanism7,8. Theseearlier studies were limited by image resolution to craters largerthan a few hundred metres in diameter. Here we report ananalysis, using new images obtained by the Mars Global Surveyorspacecraft, of crater populations that extend the size distributiondown to about 16 m. Our results indicate a wide range of surfaceages, with one regionÐlava ¯ows within the Arsia Mons calderaÐthat we estimate to be no older than 40±100 million years. Wesuggest that volcanism is a continuing process on Mars.

The distribution of crater numbers versus crater diameter onlunar lava plains, called the `production function', represents theshape of the population of craters being produced on the moon incurrent geological time. Its shape is well determined9±11. Our initialstep was to test whether the production function observed onyoung, well-preserved surfaces on Mars is the same as that foundon the Moon. This result has been found for craters larger than 1 kmin diameter, but has not been well tested for the steep branch below1 km (ref. 7) (see ®gures). Dashed reference lines in the shape of thelunar production function are shown in each of the crater countdiagrams in this Letter, along with an upper solid line that marks thecrater density on the most heavily cratered surfaces in the SolarSystem, dated about 4.0 Gyr old on the Moon, and believed to markthe saturation equilibrium condition where new craters erase oldcraters10,12. These reference lines allow the comparison of the Marscounts with the lunar production function.

Here we report our analyses of images obtained by the MarsOrbiter Camera (MOC) on board the Mars Global Surveyor space-craft. Martian crater counts obtained from the MOC images are asigni®cant advance over previously published data, and the imagesreveal the importance of mobile dust in shaping the martianlandscape and softening the pro®le of craters13,14. In general, ourprocedure is to count all craters but avoid areas with obviousclusters of small secondary ejecta craters. To study the crater sizedistribution in a relatively young area, we chose a MOC image thatcrosses a strip of the ¯oor of the summit caldera of the very youngTharsis volcano, Arsia Mons. Figure 1a shows some of this surface,and the crater counts obtained on Arsia Mons and its summitcaldera are given in Fig. 1b. The largest crater within the summitcaldera is barely 1 km across, and so the counts are extended tolarger sizes (using open symbols) with additional counts from the¯anks of Arsia Mons. These counts (caldera and whole volcano)each appear consistent with the shape of the lunar crater diameter


letters to nature

NATURE | VOL 397 | 18 FEBRUARY 1999 | www.nature.com 587

distribution, and support the contention that the recent martianproduction function matches that of the Moon at all observed sizes.At the upper left of Fig. 1b, the curve intersects the proposedsaturation equilibrium line (shown solid) at diameter D < 60 m.This behaviour is analogous to that found in the lunar maria, wherethe steep branch hits the saturation line at D < 300 m.

These data contain age information. The reference lines (Fig. 1b)show that the crater density in the summit caldera is only 2±10% ofthat found in the lunar maria; the crater density on the outer slopesmay be roughly 3±10 times that. Our interpretation is that thecaldera lavas are relatively young and that no substantial obliterativelosses have occurred for craters down to D � 60 m, or depths asshallow as ,10 m. A review12 of asteroid and cometary data andcratering physics suggests that the actual martian crater productionrate in recent geological time is 1±4 times the mean post-mare lunarrate, with a best estimate of 2. This best estimate yields an ageestimate for Arsia Mons caldera lava ¯ows of roughly 40±200 Myr,and the outer ¯ank ages would be several times older, withuncertainties of a factor of 2±3.

The good match between the slope of the martian and the lunarcrater diameter distributions, at 60 m , D , 1 km, indicates thatthere has not been enough dust in®ll in this region to remove manycraters larger than 60 m. Nonetheless, although the altitude is,26 km above the mean surface of Mars, we see direct evidencefor some dust deposition in certain areas on the caldera rim. Figure2 shows a portion of MOC image no. 3308 in a region of horst±graben structure just outside the north caldera rim of Arsia Mons.Rilles and other textures are clearly seen on the horst surfaces, butare muted or covered entirely by smooth deposits on the lowergraben ¯oors, especially in smooth drifts banked against the edges ofthe grabens. The dust source may be fallout from global dust stormsthat inject dust into layers as high as 35±40 km in the martianatmosphere15,16.

Figure 1 Crater density on Arsia Mons. a, Portion of MOC image no. 3308,

showing portion of summit caldera ¯oor on Arsia Mons. b, Crater counts for the

caldera ¯oor and ¯anks of Arsia Mons, superimposed on reference lines scaled

to crater populations on lunar lava plains. The two short solid lines represent the

stratigraphic de®nition of the division between three eras of Martian history:

Amazonian (lower part of graph), Hesperian (between the lines) and Noachian.

The counts suggest that the caldera ¯oor is younger than the ¯anks of the

volcano. (The crater counts were obtained from the images by more than one

person, to avoid bias and to test repeatability; the name of the person is given in

the key.)

Figure 2 Horst±graben structure concentric with Arsia Mons caldera rim. Parts of

graben ¯oors show evidence of dust deposits. See text for details.


letters to nature

588 NATURE | VOL 397 | 18 FEBRUARY 1999 | www.nature.com

An additional issue clari®ed by these data involves the minimumcrater size on Mars. Viking lander analysts concluded that the craterpopulation cut off below diameter D � 50 m, due to atmosphericbreakup of bolides17. The MOC images give the ®rst chance to testthat prediction. We ®nd no cut-off down to D < 16 m. Many localregions are resurfaced by dust drifts and have few small craters, butother nearby areas show old surfaces where crater numbers increasesmoothly as D decreases, down to 16 m and less.

The comparison between young areas and ancient upland areas

on Mars is striking. Figure 3a shows a moderately heavily crateredupland area near Nirgal Vallis in MOC image no. 605, and cratercounts are given in Fig. 3b. Figure 4a shows a heavily cratered terrainin MOC image no. 2303 on the ¯oor of the crater Schiaparelli, alarge, old crater which in turn is superposed on one of the mostheavily cratered martian terrains, Arabia Terra; crater counts aregiven in Fig. 4b. As is characteristic of all heavily cratered areas, thecrater counts for both these areas show a pronounced ¯attening ofthe primary crater branch from 1 km , D , 45 km. This ¯attening

Figure 3 Crater density in the area around Nirgal Vallis. a, Moderate crater density

in plains adjacent to Nirgal Vallis, in MOC image no. 605. We note the degraded

states of some craters. b, As in Fig. 1b but for the older upland region around

Nirgal Vallis, including area of image a. The solid, bent line is a calculated steady-

state line showing the OÈ pik effect for craters with constant net dust deposition of

10-6 myr-1 (W.K.H., unpublished results). The counts suggest an old surface,more

than 3Gyr old, in which smaller craters have been lost by obliterative processes,

such as dust in®ll.

Figure 4 Crater density in the area around the crater Schiaparelli. a, Portion of MOC

image no. 2303 showing heavily cratered portions of the ¯oor of the crater

Schiaparelli. Many craters are severely degraded. b, As in Fig.1b but for the ¯oor of

crater Schiaparelli and the surrounding old region of Arabia Terra, including area

of image a. The largest craters in Arabia Terra appear near saturation, and the

surface is probably ,4Gyr old. Schiaparelli appears somewhat younger, perhaps

3±4Gyr old. Smaller craters in this region haveapparently been lost byobliterative

effects; the oldest visible 20-m craters may date back no more than ,10Myr.


letters to nature

NATURE | VOL 397 | 18 FEBRUARY 1999 | www.nature.com 589

was detected by OÈ pik as early as 1965 and was attributed by him todeposition of material in craters, preferentially obliterating smallcraters1. It has subsequently been interpreted as evidence of long-term erosion, deposition, and lava ¯ooding of martian craters,especially in the earlier parts of Mars' history2±6, although it has alsobeen suggested that the early production function on Mars was¯atter than the present function18. Our MOC data show that thesteep branch of the curve in old areas also appears distinctly ¯atterthan on the Moon; in addition, the MOC images (Figs 3a and 4a)reveal a range of degradation states among 100-m-scale craters.These states range from fresh craters to craters with dune depositson the ¯oor, to craters whose ¯oors are ®lled and whose rims barelyprotrude above the dust. This ®ts the view that small craters havebeen lost by dust in®ll and blanketing, and the ¯attening of theproduction-function curves, at least at small diameters, is thusattributed to the OÈ pik effect.

The heavy, bent, solid line in Figs 3b and 4b is a predicted steady-state line for the OÈ pik effect in®lling of craters. This curve isgenerally derived in refs 3 and 4, but has been modi®ed byunpublished calculations of one of us (W.K.H.), taking into accountthe depth±diameter relation for fresh martian craters18. The averagenet deposition rate in crater ¯oors, assumed in this curve, is,10-6 m yr-1, consistent with other estimates2±7,14. The predictedcurve is a good ®t for the data. The conclusion is that on the oldestmartian uplands, smaller craters are probably in a rough equili-brium with local obliteration processes, at least if we average overlarge enough areas. A similar statement applies to Earth, but withhigher obliteration rates.

The comparison of crater size distributions on the old surfacesand young lava surfaces of Mars, and the lunar mare lava plains,indicates the wide range of surface ages on Mars, relative to theMoon; this comparison supports a conclusion that the youngestlarge-scale lava eruptions on Mars are much younger than on theMoon, having occurred in the last few per cent of martian time. Thediscovery of martian basaltic meteorites with crystallization ages of1.3 Gyr or younger19 supports this conclusion. The crater statisticsthat we report here suggest that volcanism is continuing on Mars incurrent geological time. M

Received 16 September; accepted 14 December 1998.

1. OÈ pik, E. J. Mariner IV and craters on Mars. Irish Astron. J. 7, 92±104 (1965); The Martian surface.

Science 153, 255±265 (1966).2. Hartmann, W. K. Martian cratering (Paper I). Icarus 5, 565±576 (1966).

3. Chapman, C., Pollack, J. & Sagan, C. An Analysis of the Mariner 4 Photography of Mars (Spec. Rep. 268,

Smithson. Astrophys. Obs., 1968).

4. Hartmann, W. K. Martian cratering III: Theory of crater obliterations. Icarus 15, 410±428 (1971).

5. Jones, K. L. Evidence for an episode of crater obliteration intermediate in Martian history. J. Geophys.Res. 79, 3917±3931 (1974).

6. Chapman, C. R. Cratering on Mars. I. Cratering and obliteration history. Icarus 22, 272±291 (1974).

7. Hartmann, W. K. Martian cratering, IV: Mariner 9 initial analysis of cratering chronology. J. Geophys.

Res. 78, 4096±4116 (1973).

8. Neukum, G. & Hiller, K. Martian ages. J. Geophys. Res. 86, 3097±3121 (1981).9. Strom, R. G., Croft, S. K. & Barlow, N. G. in Mars (ed. Kieffer, H.) 383±423 (Univ. Arizona Press,

Tucson, 1992).

10. Hartmann, W. K. Planetary cratering 1. Lunar highlands and tests of hypotheses on crater

populations. Meteoritics 30, 451±467 (1995).11. Plaut, J., Kahn, R., Guiness, E. & Arvidson, R. Accumulation of sedimentary debris in the south polar

region of Mars and implications for climate history. Icarus 76, 357±377 (1988).

12. Hartmann, W. K. et al. in Basaltic Volcanism on the Terrestrial Planets (eds Basaltic Volcanism Study

Project) 1050±1129 (Pergamon, Elmsford, NY, 1981).

13. Hartmann, W. K. & Gaskell, R. W. Planetary cratering 2: Studies of saturation equilibrium. Meteorit.Planet. Sci. 32, 109±121 (1996).

14. Malin, M. C. et al. Early views of the Martian surface from the Mars Orbiter camera of Mars Global

Surveyor. Science 279, 1681±1685 (1998).

15. Binder, A. B. et al. The geology of the Viking 1 lander site. J. Geophys. Res. 82, 4439±4451 (1977).

16. Gault, D. E. & Baldwin, B. S. Impact cratering on Mars: some effects in the atmosphere. Eos 51, 343(1970).

17. Carr, M. H. & Viking Orbiter Team Viking Orbiter View of Mars (Spec. Publ. 441, NASA Washington

DC, 1980).

18. Cinala, M. J. in Impact and Explosion Cratering (eds Roddy, D. J., Pepin, R. O. & Merrill, R. B.) 575±

592 (Pergamon, Elmsford, NY, 1977).19. Nyquist, L. et al. A single-crater origin for Martian shergottites: Resolution of the age paradox? Lunar

Planet. Sci. 29, 1688 (1998).

Acknowledgements. We thank G. Herres, G. Esquerdo, and, in Madrid, J. Anguita and M. de las Casas, forassistance with crater counts and data processing. We also thank D. Berman and G. Hartmann for editorialassistance.

Correspondence and requests for materials should be addressed to W.K.H. (e-mail: [email protected]).

Groundwater formationofmartianvalleysMichael C. Malin* & Michael H. Carr²

* Malin Space Science Systems, PO Box 910148, San Diego,California 92191-0148, USA² US Geological Survey, 345 Middle®eld Road, Menlo Park,

California 94025, USA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

The martian surface shows large out¯ow channels, widelyaccepted as having been formed by gigantic ¯oods that couldhave occurred under climatic conditions like those seen today1±5.Also present are branching valley networks that commonly havetributaries1±8. These valleys are much smaller than the out¯owchannels and their origins and ages have been controversial. Forexample, they might have formed through slow erosion by waterrunning across the surface, either early or late in Mars' history9±13,possibly protected from harsh conditions by ice cover14±16. Alter-natively, they might have formed through groundwater orground-ice processes that undermine the surface and causecollapse, again either early or late in Mars' history3,4. Long-duration surface runoff would imply climatic conditions quitedifferent from the present environment. Here we present high-resolution images of martian valleys that support the view thatground water played an important role in their formation,although we are unable as yet to establish when this occurred.

Images acquired by the Mars Orbiter Camera (MOC) during theaerobraking phase (September 1997 to February 1998) of the MarsGlobal Surveyor mission typically have resolutions in the range 4±8 m per pixel, in most cases a factor of 20±50 times better thanprevious imaging17,18. The images reveal new details about thevalleys that strongly support an origin by ¯uid erosion. Althoughapparent drainage networks are observed locally, dissection of theadjacent upland surface, as might be expected if the ¯uid had anatmospheric rather than a subsurface source, is not seen. The lack oferoded uplands adjacent to martian ¯uvial valleys and the implica-tion for a localized water source was previously noted in studies ofViking Orbiter images19,20.

Figure 1 is an image of Nanedi VallisÐan 800-km-long valleythat appears incised into cratered plains north of the VallesMarineris. Nanedi Vallis is one of the longest and freshest-appearingof the martian valley networks. Despite its length it has only a fewshort tributaries, and no obvious catchment area. It starts close tothe equator at 498 W in an area where there is other evidence forgroundwater action, including the source of the out¯ow channel,Shalbatana Vallis. The circuitous path of the valley seen here appearsto have been inherited from sur®cial ¯uid movement, although thesource of the ¯uid is not apparent. Such arcuate and reversing pathsare dif®cult, if not impossible, to create by headward erosion (thatis, progressively upstream towards the source) of a stream, lendingsupport to an interpretation, based on visual appearance, that thevalley formed by entrenchment of an originally meandering ¯ow.This interpretation is further strengthened by the observation of aninterior channel, presumably the speci®c course of the valley-forming ¯uid. However, as with many entrenched valley systemson Earth, mass movements accompanying groundwater action arelikely to have created much of the relief and width of the presentlyobserved valley. It could be argued that valley formation re¯ectedgroundwater processes fed by precipitation, and that the lack ofdissection of the adjacent plain is the result of high permeability ofthe near-surface materials21. However, the total absence of metre-scale dissection here and elsewhere on Mars, and the near-absence ofupstream tributaries (suggesting a spatially limited source), supporta subsurface rather than atmospheric source.

These MOC images show clearly that the uncratered and often

ExoMarsSearching for Life on the Red Planet

ExoMarsSearching for Life on the Red Planet

E stablishing whether life ever existed onMars, or is still active today, is an

outstanding question of our time. It is alsoa prerequisite to prepare for future humanexploration. To address this importantobjective, ESA plans to launch the ExoMarsmission in 2011. ExoMars will also develop anddemonstrate key technologies needed to extendEurope’s capabilities for planetary exploration.

Mission ObjectivesExoMars will deploy two science elementson the Martian surface: a rover and asmall, fixed package. The Rover willsearch for signs of past and present lifeon Mars, and characterise the water andgeochemical environment with depth bycollecting and analysing subsurfacesamples. The fixed package, theGeophysics/Environment Package (GEP),will measure planetary geophysics para-meters important for understandingMars’s evolution and habitability,identify possible surface hazards tofuture human missions, and study theenvironment.

The Rover will carry a comprehensivesuite of instruments dedicated to exo-biology and geology: the Pasteur payload.It will travel several kilometres searchingfor traces of life, collecting andanalysing samples from inside surfacerocks and by drilling down to 2 m. Thevery powerful combination of mobilityand accessing locations where organicmolecules may be well-preserved isunique to this mission.

Jorge Vago, Bruno Gardini, Gerhard Kminek,Pietro Baglioni, Giacinto Gianfiglio,Andrea Santovincenzo, Silvia Bayón& Michel van WinnendaelDirectorate for Human Spaceflight,Microgravity and Exploration Programmes,ESTEC, Noordwijk, The Netherlands

esa bulletin 126 - may 2006 17

ExoMars

ExoMars will also pursue importanttechnology objectives aimed at extend-ing Europe’s capabilities in planetaryexploration. It will demonstrate thedescent and landing of a large payload onMars; the navigation and operation of amobile scientific platform; a novel drill toobtain subsurface samples; and meetchallenging planetary protection andcleanliness levels necessary to achieve themission’s ambitious scientific goals.

The Search for LifeExobiology, in its broadest terms, denotesthe study of the origin, evolution anddistribution of life in the Universe. It iswell established that life arose very earlyon the young Earth. Fossil records showthat life had already attained a large degreeof biological sophistication 3500 millionyears ago. Since then, it has provedextremely adaptable, colonising even themost disparate ecological habitats, fromthe very cold to the very hot, and spanninga wide range of pressure and chemicalconditions. For organisms to have emergedand evolved, water must have been readilyavailable on our planet. Life as we know itrelies, above all else, upon liquid water.Without it, the metabolic activities ofliving cells are not possible. In the absenceof water, life either ceases or slips intoquiescence.

Mars today is cold, desolate and dry. Itssurface is highly oxidised and exposed tosterilising and degrading ultraviolet (UV)radiation. Low temperature and pressurepreclude the existence of liquid water;except, perhaps, in localised environments,and then only episodically. Nevertheless,numerous features such as large channels,dendritic valley networks, gullies and

sedimentary rock formations suggest thepast action of surface liquid water on Mars– and lots of it. In fact, the sizes of out-flow channels imply immense discharges,exceeding any floods known on Earth.

Mars’s observable geological recordspans some 4500 million years. From the number of superposed craters, the oldest terrain is believed to be about4000 million years old, and the youngestpossibly less than 100 million years.Most valley networks are ancient(3500–4000 million years), but as manyas 25–35% may be more recent. Today,water on Mars is only stable as ice at thepoles, as permafrost in widespreadunderground deposits, and in traceamounts in the atmosphere. From abiological perspective, past liquid wateritself motivates the question of life onMars. If Mars’ surface was warmer andwetter for the first 500 million years ofits history, perhaps life arose independ-ently there at more or less the same timeas it did on Earth.

An alternative pathway may have beenthe transport of terrestrial organismsembedded in meteoroids, delivered fromEarth. Yet another hypothesis is that lifemay have developed within a warm, wetsubterranean environment. In fact, giventhe discovery of a flourishing biosphere akilometre below Earth’s surface, a similarvast microbial community may be activeon Mars, forced into that ecological nicheby the disappearance of a more benignsurface environment. The possibility thatlife may have evolved on Mars during anearlier period surface water, and thatorganisms may still exist underground,marks the planet as a prime candidate inthe search for life beyond Earth.

Hazards for Manned Operations on MarsBefore we can contemplate sendingastronauts to Mars, we must understandand control any risks that may pose athreat to a mission’s success. We can begin to assess some of these risks withExoMars.

Ionising radiation is probably the singlemost important limiting factor for human interplanetary flight. To evaluateits danger and to define efficient mitigationstrategies, it is desirable to incorporateradiation-monitoring capabilities duringcruise, orbit and surface operations onprecursor robotic missions to Mars.

Another physical hazard may resultfrom the basic mechanical properties ofthe Martian soil. Dust particles will invade the interior of a spacecraft during surface operations, as shownduring Apollo’s operations on the Moon. Dust inhalation can pose a threat to astronauts on Mars, and evenmore so under microgravity during thereturn flight to Earth. Characteristics ofthe soil, including the sizes, shapes andcompositions of individual particles,can be studied with dedicated in situinstrumentation. However, a more in-depth assessment, including a toxicity analysis, requires the return of asuitable Martian sample.

Reactive inorganic substances couldpresent chemical hazards on the surface.Free radicals, salts and oxidants are veryaggressive in humid conditions such as thelungs and eyes. Toxic metals, organics andpathogens are also potential hazards. Aswith dust, chemical hazards in the soil willcontaminate the interior of a spacecraftduring surface operations. They coulddamage the health of astronauts and the

esa bulletin 126 - may 2006 www.esa.int18

operation of equipment. Many potentialinorganic and organic chemical hazardsmay be identified with the ExoMarssearch-for-life instruments.

Geophysics MeasurementsThe processes that have determined thelong-term ‘habitability’ of Mars dependon the geodynamics of the planet, and on its geological evolution and activity.Important issues still need to be resolved.

What is Mars’ internal structure? Is thereany volcanic activity on Mars? Theanswers may allow us to extrapolate intothe past, to estimate when and how Mars lost its magnetic field, and theimportance of volcanic outgassing for theearly atmosphere.

ExoMars will also carry the Geophysics/Environment Package, accommodated onthe Descent Module and powered by asmall radioisotope thermal generator.

Searching for Signs of LifeIf life ever arose on the Red Planet, itprobably did so when Mars was warmerand wetter, during its initial 500–1000 million years. Conditions then weresimilar to those on early Earth: activevolcanism and outgassing, meteoriticimpacts, large bodies of liquid water, anda mildly reducing atmosphere. We mayreasonably expect that microbes quicklybecame global. Nevertheless, there isinevitably a large measure of chanceinvolved in finding convincing evidence ofancient, microscopic life forms.

On Earth’s surface, the permanentpresence of running water, solar-UVradiation, atmospheric oxygen and lifeitself quickly erases all traces of anyexposed, dead organisms. The onlyopportunity to detect them is to find theirbiosignatures encased in a protectiveenvironment, as in suitable rocks.However, since high-temperature meta-morphic processes and plate tectonicshave reformed most ancient terrains, it isvery difficult to find rocks on Earth olderthan 3000 million years in good condition.Mars, on the other hand, has not sufferedsuch widespread tectonic activity. Thismeans there may be rock formations fromthe earliest period of Martian history thathave not been exposed to high-temperaturerecycling. Consequently, well-preservedancient biomarkers may still be accessiblefor analysis.

Even on Earth, a major difficulty insearching for primitive life is that, inessence, we are looking for the remnants ofminuscule beings whose fossilised formscan be simple enough to be confused withtiny mineral precipitates. This issue lies atthe heart of a heated debate among

esa bulletin 126 - may 2006www.esa.int 19

The ExoMars Rover will be able to drill down to 2 m for samples

A Mars Express image of the Ares Vallis region,showing evidence of ancient, vast water discharges. This immense channel, 1400 km long, empties intoChryse Planitia, where Mars Pathfinder landed in1997. (ESA/DLR/FU Berlin, G. Neukem)

Aurora


Mission strategy to achieve ExoMars’s scientific objectives:

1 To land on, or be able to reach, a location with high exobiology interest for past and/orpresent life signatures, i.e. access to the appropriate geological environment.

2 To collect scientific samples from different sites, using a Rover carrying a drill capable ofreaching well into the soil and surface rocks. This requires mobility and access to thesubsurface.

3 At each site, to conduct an integral set of measurements at multiple scales: beginning witha panoramic assessment of the geological environment, progressing to smaller-scaleinvestigations on interesting surface rocks using a suite of contact instruments, andculminating with the collection of well-selected samples to be studied by the Rover’sanalytical laboratory.

4 To characterise geophysics and environment parameters relevant to planetary evolution,life and hazards to humans.

To arrive at a clear and unambiguous conclusion on the existence of past or present life atthe Rover sites, it is essential that the instrumentation can provide mutually reinforcing lines ofevidence, while minimising the opportunities for alternative interpretations.

It is also imperative that all instruments be carefully designed so that none is a weak link inthe chain of observations; performance limitations in an instrument intended to confirm theresults obtained by another should not generate confusion and discredit the wholemeasurement.

The science strategy for the Pasteur payload is therefore to provide a self-consistent set ofinstruments to obtain reliable evidence, for or against, the existence of a range ofbiosignatures at each search location.

Spacecraft: Carrier plus Descent Module (including Rover and GEP)Data-relay provided by NASA

Launch: May–June 2011, from Kourou on Soyuz-2b (backup 2013)

Arrival: June 2013 (backup 2015)

Landing: Direct entry, from hyperbolic trajectory, after the dust storm season. Latitudes 15˚S–45˚N, all longitudes, altitude: <0 m, relative to the MGS/MOLA* zero level

Science: Rover with Pasteur payload:mass 120–180 kg, includes: Drill System/SPDS and instruments (8 kg); lifetime 180 sols

Geophysics Environment Package (GEP):mass <20 kg; includes: instruments (~4 kg); lifetime 6 years

Ground Mission control and mission operations: ESOCSegment: Rover operation on Mars surface: Rover Operations Centre

GEP operations: to be decided

*MGS/MOLA: Mars Global Surveyor/Mars Orbiter Laser Altimeter

palaeobiologists. It is therefore doubtfulthat any one signature suggestive of life– whether it is an image implying abiostructure, an interesting organiccompound or a fractionated isotopic ratio– may reliably demonstrate a biogenicorigin. Several independent lines ofevidence are required to construct acompelling case. ExoMars must thereforepursue a holistic search strategy, attackingthe problem from multiple angles,including geological and environmentalinvestigations (to characterise potentialhabitats), visible examination of samples(morphology) and spectrochemicalcomposition analyses.

In 1976, the twin Viking landersconducted the first in situ measurementsfocusing on the detection of organiccompounds and life on Mars. Theirbiology package contained threeexperiments, all looking for signs ofmetabolism in soil samples. One, thelabelled-release experiment, producedprovocative results. If other informationhad not been also available, these datacould have been interpreted as proof ofbiological activity. However, theoreticalmodelling of the atmosphere and regolithchemistry hinted at powerful oxidantsthat could more-or-less account for theresults of the three experiments. Thebiggest blow was the failure of the Vikinggas chromatograph mass spectrometer(GCMS) to find evidence of organicmolecules at the parts-per-billion level.

With few exceptions, the majority of thescientific community has concluded thatthe Viking results do not demonstrate thepresence of life. Numerous attempts havebeen made in the laboratory to simulatethe Viking reactions. While some havereproduced certain aspects, none hassucceeded entirely. Incredibly, 30 yearsafter Viking, the crucial chemical oxidanthypothesis remains untested. ExoMarswill include a powerful instrument tostudy oxidants and their relation toorganics distribution on Mars.

Undoubtedly, the present environmenton Mars is exceedingly harsh for thewidespread proliferation of surface life: itis simply too cold and dry, not to mentionthe large doses of UV. Notwithstanding

Recommended Pasteur Exobiology Instruments1

PPaannoorraammiicc To characterise the Rover’s geological context (surface and subsurface). Typical scales span from panoramic to IInnssttrruummeennttss 10 m, with a resolution of the order of 1 cm for close targets.

Panoramic Camera 2 wide-angle stereo cameras and 1 high-resolution camera; to characterise the Rover’s environment and itsSystem geology. Also very important for target selection.

Infrared (IR) For the remote identification of water-related minerals, and for target selection.Spectrometer

Ground Penetrating) To establish the subsurface soil stratigraphy down to 3 m depth, and to help plan the drilling strategy.Radar (GPR)

CCoonnttaacctt To investigate exposed bedrock, surface rocks and soils. Among the scientific interests at this scale are:IInnssttrruummeennttss macroscopic textures, structures and layering; and bulk mineralogical and elemental characterisation. This

information will be fundamental to collect samples for more detailed analysis. The preferred solution is todeploy the contact instruments using an arm-and-paw arrangement, as in Beagle-2. Alternatively, in case of masslimitations, they could be accommodated at the base of the subsurface drill.

Close-Up Imager To study rock targets visually at close range (cm) with sub-mm resolution.

Mössbauer Spectrometer To study the mineralogy of Fe-bearing rocks and soils.

Raman-LIBS2 external To determine the geochemistry/organic content and atomic composition of observed minerals. These opticalheads are external heads connected to the instruments inside the analytical laboratory.

SSuuppppoorrtt These instruments are devoted to the acquisition and preparation of samples for detailed investigations in theIInnssttrruummeennttss analytical laboratory. They must follow specific acquisition and preparation protocols to guarantee the optimal

survival of any organic molecules in the samples. The mission’s ability to break new scientific ground, particularlyfor signs-of-life investigations, depends on these two instruments.

Subsurface Drill Capable of obtaining samples from 0 m to 2 m depths, where organic molecules might be well-preserved. It also integrates temperature sensors and an IR spectrometer for borehole mineralogy studies.

Sample Preparation Receives a sample from the drill system, prepares it for scientific analysis, and presents it to all analytical and Distribution laboratory instruments. A very important function is to produce particulate material while preserving the System (SPDS) organic and water content.

AAnnaallyyttiiccaall To conduct a detailed analysis of each sample. The first step is a visual and spectroscopic inspection. If theLLaabboorraattoorryy sample is deemed interesting, it is ground up and the resulting particulate material is used to search for organic

molecules and to perform more accurate mineralogical investigations.

Microscope IR To examine the collected samples to characterise their structure and composition at grain-size level. These measurements will also be used to select sample locations for further detailed analyses by the Raman-LIBS spectrometers.

Raman-LIBS To determine the geochemistry/organic content and elemental composition of minerals in the collected samples.

X-ray Diffractometer (XRD) To determine the true mineralogical composition of a sample’s crystalline phases.

Urey (Mars Organics Mars Organics Detector (MOD): extremely high-sensitivity detector (ppt) to search for amino acids, nucleotide and Oxidants Detector) bases and PAHs in the collected samples. Can also function as front-end to the GCMS. Mars Oxidants

Instrument (MOI): determines the chemical reactivity of oxidants and free radicals in the soil and atmosphere.

GCMS Gas chromatograph mass spectrometer to conduct a broad-range, very-high sensitivity search for organic molecules in the collected samples; also for atmospheric analyses.

Life-Marker Chip Antibody-based instrument with very high specificity to detect present life reliably.

1Mass (without drill and SPDS): 12.5 kg. 2LIBS: Laser-Induced Breakdown Spectroscopy.

Recommended Pasteur Environment Instruments3

EEnnvviirroonnmmeenntt To characterise possible hazards to future human missions and to increase our knowledge of the Martian IInnssttrruummeennttss environment.

Dust Suite Determines the dust grain size distribution and deposition rate. It also measures water vapour with high precision.

UV Spectrometer Measures the UV radiation spectrum.

Ionising Radiation Measures the ionising radiation dose reaching the surface from cosmic rays and solar particle events.

Meteorological Package Measures pressure, temperature, wind speed and direction, and sound.

3Mass: 1.9 kg. The Pasteur environment instruments are presently planned to be accommodated in the GEP.


ExoMars

these hazards, basic organisms could stillflourish in protected places: deepunderground, at shallow depths inespecially benign environments, or withinrock cracks and inclusions.

The strategy to find traces of pastbiological activity rests on the assumptionthat any surviving signatures of interestwill be preserved in the geological record,in the form of buried/encased remains,organic material and microfossils.Similarly, because current surfaceconditions are hostile to most knownorganisms, as when looking for signs ofextant life, the search methodology shouldfocus on investigations in protectedniches: underground, in permafrost orwithin surface rocks. This means thatthere is a good possibility that the samesampling device and instrumentationmay adequately serve both types ofstudies. The biggest difference is due tolocation requirements. In one case, theinterest lies in areas occupied by ancientbodies of water over many thousands of years. In the other, the emphasis is onwater-rich environments close to thesurface and accessible to our sensorstoday. For the latter, the presence ofpermafrost alone may not be enough.Permafrost in combination with asustained heat source, probably of volcanicor hydrothermal origin, may be necessary.Such warm oases can only be identified byan orbital survey of the planet. In the next

few years, a number of remote-sensingsatellites, like ESA’s Mars Express andNASA’s Mars Reconnaissance Orbiter(MRO), will determine the water/iceboundary across Mars and may help todiscover such warm spots. If they do exist,they would be prime targets for missionslike ExoMars.

On Earth, microbial life quickly becamea global phenomenon. If the sameexplosive process occurred on the youngMars, the chances of finding evidence of itare good. Even more interesting would bethe discovery and study of life forms thathave successfully adapted to the modernMars. However, this presupposes the prioridentification of geologically suitable, life-friendly locations where it can bedemonstrated that liquid water still exists,at least episodically throughout the year.For these reasons, the ‘Red Book’ scienceteam advised ESA to focus on thedetection of extinct life, but to buildenough flexibility into the mission to beable to target sites with the potential forpresent life.

Mission DescriptionThe baseline mission scenario consists ofa spacecraft composite with a Carrier anda Descent Module, launched by aSoyuz-2b from Kourou. It will follow a2-year ‘delayed trajectory’ in order toreach Mars after the dust-storm season.The Descent Module will be released from

the hyperbolic arrival path, and landusing either bouncing (non-vented, as inNASA’s rovers) or non-bouncing (vented)airbags, and deploy the Rover and GEP.In the baseline mission, data-relay forthe Rover will be provided by a NASAorbiter.

An alternative configuration, based onan Ariane-5 ECA launcher, may beimplemented depending on programme,technical and financial considerations.In this option, the Carrier is replacedwith an Orbiter that provides end-to-enddata relay for the surface elements. TheOrbiter will also carry a science payloadto complement the results from theRover and GEP, and provide continuityto the great scientific discoveries flowingfrom Mars Express.

ExoMars is a search-for-life missiontargeting regions with high life potential.It has therefore been classified asPlanetary Protection category IVc. This,coupled with the mission’s ambitiousscientific goals, imposes challengingsterilisation and organic cleanlinessrequirements.

The ExoMars RoverThe Rover will have a nominal lifetime of180 sols (about 6 months). This periodprovides a regional mobility of severalkilometres, relying on solar array electricalpower. The Pasteur model payloadincludes panoramic instruments (cameras,ground-penetrating radar and IR spectro-meter; contact instruments for studyingsurface rocks (close-up imager andMössbauer spectrometer), a subsurfacedrill to reach depths of 2 m and tocollect specimens from exposed bedrock,a sample preparation and distributionunit, and the analytical laboratory. Thelatter includes a microscope, anoxidation sensor and a variety of

Aurora


The ExoMars surface science exploration scenario. The Rover willconduct measurements of multiple scales, starting with apanoramic assessment of the geological environment, progressingto more detailed investigations on surface rocks using a suite ofcontact instruments, and culminating with the collection of well-selected samples to be analysed in its laboratory

instruments for characterising theorganic substances and geochemistry inthe collected samples.

A key element is the drill. The reason forthe 2 m requirement is the need to obtainpristine sample material for analysis.Whereas the estimated extinction horizonfor oxidants in the subsurface is severalcentimetres, damaging ionising radiationcan penetrate to depths of around 1 m.Additionally, it is unlikely that loose dustmay hold interesting biosignatures,because it has been moved around bywind and processed by UV radiation. Inthe end, organic substances may best bepreserved within low-porosity material.Hence, the ExoMars drill must be ableto penetrate and obtain samples fromwell-consolidated (hard) formations,such as sedimentary rocks and evaporiticdeposits. Additionally, it must monitorand control torque, thrust, penetrationdepth and temperature at the drill bit.Grain-to-grain friction in a rotary drillcan generate a heat wave in the sample,

destroying the organic molecules thatExoMars seeks to detect. The drill musttherefore have a variable cutting protocol,to dissipate heat in a science-safemanner. Finally, the drill’s IR spectro-meter will conduct mineralogy studiesinside the borehole.

ConclusionNASA’s highly successful 2004 roverswere conceived as robotic geologists. Theyhave demonstrated the past existence oflong-lasting, wet environments on Mars.Their results have persuaded the scientificcommunity that mobility is a must-haverequirement for all future surface missions.Recent results from Mars Express haverevealed multiple, ancient depositscontaining clay minerals that form only in the presence of liquid water. Thisreinforces the hypothesis that ancientMars may have been wetter, and possiblywarmer, than it is today. NASA’s 2009Mars Science Laboratory will studysurface geology and organics, with the

goal of identifying habitable environ-ments. ExoMars is the next logical step.It will have instruments to investigatewhether life ever arose on the RedPlanet. It will also be the first missionwith the mobility to access locationswhere organic molecules may be well-preserved, thus allowing, for the firsttime, investigation of Mars’ thirddimension: depth. This alone is aguarantee that the mission will breaknew scientific ground. Finally, the manytechnologies developed for this projectwill allow ESA to prepare forinternational collaboration on futuremissions, such as Mars Sample Return.

Following the recent accomplishmentsof Huygens and Mars Express,ExoMars provides Europe with a newchallenge, and a new opportunity todemonstrate its capacity to performworld-class planetary science.

ExoMars is now in Phase-B1 and isexpected to begin Phase-B2 in mid-2007and Phase-C/D in early 2008. e


ExoMars

The analysis sequence within Pasteur’s analytical laboratory

The Pasteur payload’s drill-bit design concept. The drill’s full 2 mextension is achieved by assembling four sections (one drill toolrod, with an internal shutter and sample-collection capability, plusthree extension rods). The drill will also be equipped with an IRspectrometer for mineralogy studies inside the borehole. (GalileoAvionica)

Doctoral Dissertation Research Proposal:Geographic Representations of the Planet Mars, 1867-1907

K. Maria D. LaneDepartment of Geography, University of Texas

Committee: Ian R. Manners (Chair), Diana K. Davis,Steven D. Hoelscher, Kelley A. Crews-Meyer, Roger Hart

Abstract

This dissertation research will use archival and interpretive methods to examinegeographical representations of the planet Mars produced by Western astronomers and sciencewriters in the late nineteenth century. Specifically, this project will investigate the ways in whichthe development of cartography and texts portraying Mars between 1867 and 1907 participated inwider ideological discourses concerning science, imperialism and modernity. The workinghypothesis is that representations of Mars’ geography not only reflected the social contexts ofastronomical societies, sponsored observatories, and the larger Western scientific communitiesthrough the use of common textual tropes and cartographic conventions; but that they also servedto modify or construct these very contexts. The proposed research will investigate the extent towhich historical geographies of Mars challenged or altered dominant discourses of modernWestern superiority by representing the planet as a landscape inhabited by beings with superiorengineering and organizational skills.

This inquiry will be conducted through archival investigation of three specific conflicts inthe representation of Mars that marked turning points in the planet’s astronomy: over (1) thenomenclature assigned to its geographical features, (2) the mapping of canals on its surface, and(3) the interpretation of such canals as the work of intelligent beings. Interpretive analysis ofarchival materials – including astronomers’ original maps, sketches, manuscripts, observationlogbooks, correspondence and lectures; popular media coverage of astronomers’ findings; andcontemporaneous maps and documents produced for imperial and other purposes – will focus onreconstructing the historical, social and cultural contexts in which astronomers worked, while alsoestablishing the extent to which discourses of Mars’ geography infiltrated other scientific,imperial and popular dialogues during the same time period. Analysis will be guided by thehypothesis stated above, but will remain open to other scientific-cultural explanations for thenature and meaning of what appear today to be rather curious and remarkable geographies of theMartian landscape. By focusing on maps and texts that are relatively unknown to scholarsoutside the history of astronomy, this research will contribute materially and theoretically to thehistory of cartography, science studies, historical geography and studies of colonialism.

Introduction and Research Question

Despite the growing interest in nineteenth-century geographical representation, no

geographer has yet seriously examined the remarkable discourses that emerged during the latter

half of the century to represent the geography of worlds beyond Earth. Popular histories of

astronomy (e.g. Sheehan 1996; Morton 2002) indicate that astronomers collected extensive

geographic data about the nearby planets, usually recording their findings in detailed maps that

were strikingly similar in appearance to many of the well-studied imperial maps produced during

the same time period. Although much of this astronomical-geographical knowledge compiled

during the late nineteenth century has since been revised or discarded on the basis of twentieth-

century remote sensing images, I contend that colonial-era discourses concerning otherworldly

geographies had widespread scientific and cultural significance at the time they were created.

The representation of Mars as a canal-covered landscape in 1877, for example, not only

reverberated throughout the Western world’s scientific communities, but also initiated a storm of

public debate and speculation regarding humankind’s isolation in the universe. Numerous

astronomers’ claims that they could see a canal network on the Martian surface induced

widespread theoretical acceptance of the “plurality of worlds” (the existence of humanoid life on

celestial bodies other than Earth) in both Europe and the United States over nearly four decades

(Guthke 1983; Crowe 1986). Despite the clear cultural significance of this episode, it has largely

been dismissed by standard teleological approaches in the history of astronomy as a case of

scientific error. The proposed research rejects that interpretation, suggesting instead that a

detailed investigation of the statements and interactions of individual astronomers, scientists,

public officials and even public media between 1867 and 1907 will reveal the Mars “canal craze”

to be a significant and complex negotiation of sciences, cultures, and modernities. Specifically,

the project will address the following research questions:

2

1. In what ways did prominent nineteenth-century geographical discourses regarding Mars’surface features and inhabitants reflect the specific social contexts of astronomical societies,sponsored observatories, and the larger Western scientific communities?

2. How were scientific representations of Mars as an inhabited, irrigated planet contested and,ultimately, widely accepted as true in Europe and the United States?

3. To what extent did geographies of Mars challenge dominant discourses of modern Westernsuperiority by representing the planet as a landscape inhabited by beings with superiorengineering and organizational skills?

Theoretical Context

This project is theoretically informed by several related literatures that form a compelling

interdisciplinary intersection: studies of colonialism, the history of cartography, and science

studies. The proposed project will draw from recent inquiries in these literatures, contributing

materially or theoretically to each.

Studies of Colonialism

Historically, the late nineteenth-century production of scientific Mars maps coincided

with a period of intense European imperialism, during which both science and cartography

(especially scientific cartography) were fundamental to the establishment and maintenance of

European power in the colonial realms (Anderson 1991; Godlewska 1995; Ryan 1996; Edney

1997). In the last two decades, studies of nineteenth-century imperialism and colonialism have

been dominated by post-colonial scholarship that concerns itself with analysis of the ways in

which imperial (and, to a lesser extent, indigenous) maps and texts constitute “discourses,”

through which knowledge and power have been negotiated and institutionalized in various

regions of the world.

The foundational post-colonial work, Said’s Orientalism (1978), argued that imperialism

depended for its power on discursive strategies and social practices that constructed geographical

knowledge about the colonial realm. Specifically, Said analyzed “Orientalist” discourse to show

that Western geographic knowledge about the Islamic world has relied on implicit epistemologies

that powerfully support Western dominance of Islamic regions and peoples. He claimed that

3

Western Orientalists’ creation of an “imaginative geography” to describe the Islamic world is

traceable in the repetition of certain tropes and literary conventions, and that the uncritical

acceptance and repetition of these tropes and conventions in Orientalist scholarship frequently

resulted in an imaginative discourse that bore little relation to the region’s actual geography.

According to Said, Orientalist writing should thus be viewed less as a commentary on the Orient

itself than as a reflection of the Occident, showing that Europeans establish their identity in

opposition to non-Europeanness, establishing themselves always in a superior, hegemonic

position.

Despite its merits, however, Said’s work has rightly been criticized for presenting an

essentialized, totalizing view of Western scholarship. Although Orientalism analyzed individual

texts and authors, Said painted them as powerfully constrained within the bounds of Orientalist

structure and ignored any resistance or divergence of approaches, thus leaving himself open to

damaging criticism that his analysis implicated all Western authors/scholars in the production of

imaginative geographies that fueled imperialism (Driver 1992). Subsequent post-colonial

scholarship has helpfully focused its scope on a wider variety of Orientalist texts, authors, genres

and historical situations (see especially Lowe 1991; Pratt 1992), highlighting many cases of

discursive resistance to imperial ideologies and activities.

The proposed research will contribute to colonial studies not only by analyzing the extent

to which selected astronomical imagery and writing served to construct a previously unstudied

“imaginative geography” of Mars, but also by assessing the possibility that such representation

constituted a challenge to the dominant Orientalism Said identified. Preliminary analysis

suggests that nineteenth-century astronomers and popular science writers used common tropes

and metaphors to make the planet’s unfamiliar geography conceptually accessible and familiar to

scientific colleagues and popular audiences. Through repetition and uncritical citation of each

other’s work, it appears that European and American astronomers created a powerful discourse

that represented the red planet as an Earthlike, inhabited, engineered, and irrigated landscape.

4

This discourse employed a number of familiar metaphors that were also present in orientalist and

colonial texts, including association with the eternal and immutable classical world (Godlewska

1995), the supposed crippling aridity (Saberwal 1997; Grove 1997) and ruined landscape (Grove

and Rackham 2001) of the distant realm, and environmental determinism of inhabitants’

physique/intelligence (Hudson 1977). The nineteenth-century imaginative geography of Mars,

like those produced by Orientalists to represent the Islamic world, was certainly more reflective

of astronomers’ own geographical notions than of the reality of Mars’ surface characteristics.

Nonetheless, it seems to have constrained subsequent investigations and compelled certain

perspectives of Mars’ geography until at least the 1960s, when photographic imagery taken by

remote probes contradicted the view of Mars as an inhabited planet.

Interestingly, however, the standard imaginative representations of Mars appear to have

departed from or challenged several well-known imperial tropes, including the presentation of the

unknown realm as an empty wilderness (Blaut 1993), the effacement of human presence (Pratt

1992) or “creative destruction” of an existing culture to make way for European customs

(Godlewska 1995), the presentation of any inhabitants as backward and depraved (Said 1978),

and the assumed superiority of European civilization through technology (Godlewska 1995). The

imagined Martians of the late nineteenth century were not the animal-like inhabitants that

Europeans described after visiting the Orient; they were skilled, noble engineers who managed to

irrigate their arid planet with a massive global system of interlinked canals. These cosmic

neighbors were hardly inferior to the modern European technologists who had just completed

their first major canal (Suez) in 1869.

This preliminary analysis raises exciting challenges to Said’s widely-accepted concept of

Orientalism, suggesting that the discourse astronomers engaged in to represent the geography of

Mars constructed a familiar imaginative Other that was actually superior to modern Europeans.

The ways in which conceptual engagement with this Other – through scientific, philosophical,

5

and popular discourses – may have deflected, challenged or transformed modernity’s truth claims

in the West will serve as the primary inquiry of the proposed research.

History of Cartography

Recent scholarship in the history of cartography has paralleled and contributed to many

of the developments described above in studies of colonialism. Rejecting a common view of the

history of cartography as a linear evolution of cartographic styles, conventions and technologies

that continually increase the accuracy of observation, calculation and representation in

mapmaking, new approaches in the last few decades have begun to question this prevailing

narrative’s Eurocentrism, its claims to objective truth, and its linear notions of progress. A

number of powerful revisions have instead begun to examine cartography as a cultural practice,

fraught with ideological meanings and distortions that undermine its claims to scientific

objectivity (Edgerton 1987; Boelhower 1988; Harley 1989). Following Harley’s (1989)

revolutionary contention that maps should be read as ideological, cultural texts, cartography has

generally come to be accepted as a form of discourse, in which knowledge and power are

expressed and negotiated.

Given cartography’s long association with the exercise of political and military power,

Harley suggested that geographers must consider how the particular historical and ideological

circumstances of a map’s production, use and consumption reflect and establish such power

(Harley 1988). Accordingly, some of the most productive recent work in the history of

cartography has critically examined map series prepared by colonial-era explorers and

administrators, especially examining the ways in which imperial cartographies metaphorically

justified colonial activities or erased indigenous peoples from desirable territories (Ryan 1994;

Carter 1999; Edney 1997). These works indicate that even reconnaissance cartographies

representing basic geographic data necessarily carry ideological meaning. The identity and

ideological position of the map’s maker, patron, and audience have been shown to fundamentally

6

influence the ways in which maps operate to construct or limit geographical knowledge

(Helgerson 1988; Ryan 1994; Manners 1997).

This literature clearly informs the proposed examination of nineteenth-century Mars

maps (and related representations), requiring a critical evaluation of the social contexts of

astronomical societies and sponsored observatories in which individual Mars astronomers

worked. Additionally, however, I suggest that the proposed research will contribute materially to

the history of cartography by bringing to light a series of map, imagery and texts that closely

paralleled the well-studied imperial maps of the same era. As argued above, the ways in which

Mars representations differed from standard imperial representations of unknown territories

should be examined as a potential site of resistance, in which astronomers differentiated

themselves from surveyors or geographers and presented a very different view of Western

superiority.

Although historians of astronomy and popular science biographers have made some use

of the collections I intend to visit, I argue that no modern researcher has seriously considered the

turn-of-the-century Mars maps as meaningful scientific or cartographic documents. Despite the

fact that historians and geographers have extensively examined imperial cartography produced in

the same era, the Mars canal maps have largely been disregarded as curiosities. The few

historical works to examine the collections that are relevant to this research have either limited

their analyses to chronological accounts of observation technology (Sheehan 1996) and the

origins of present-day Mars nomenclature (Blunck 1977), or have dealt only briefly with

nineteenth-century mapping as a historical backdrop to analysis of the cultural impact of

twentieth-century photographic Mars maps (Morton 2002). By focusing on Mars maps, drawings

and related items (correspondence, publications, lectures, etc.) that are relatively unknown to

critical scholars outside the history of astronomy, this research will provide a material

contribution to the history of cartography.

7

Science Studies

The investigative framework for this study is informed primarily by recent advances in

science studies. Popular constructivist assertions of the 1970s and 1980s (e.g. Collins 1974;

Callon 1986) – that scientific knowledge is influenced by a cultural dimension – have been

replaced by critical scholarship that has formulated a model of science as culture (Shapin and

Schaffer 1985; Biagioli 1993). In re-reading and revising historical accounts that treat science

and culture as separate entities, this new approach to science studies suggests that scientific

change occurs as a result of complex cultural negotiation. Identifying numerous instances of

“translation” and “hybridization” (subversive appropriation) of one culture’s knowledge/power

claims by another, recent scholarship rejects Kuhn’s (1970) idea that science proceeds in

revolutionary leaps and bounds.

Contemporary science studies has engaged concepts developed in cultural studies to

explain the nature of scientific practice and its knowledge/power claims. A linguistic- or

discourse-based approach to the ways in which science is negotiated and formulated with various

cultural practices or political ideologies, however, raises a host of complex new issues for

historians of science to contend with. In critiquing the universal view of science, for instance,

historians of science must avoid lapsing into relativistic accounts of the linguistic

“incommensurability” of cultural worldviews or knowledge claims (Hart 1999). Such relativism

problematically asserts an unbridgeable divide between the “West” and its “Others,” lending

unfortunate credence to the persistent notion that Westerners have achieved superiority over non-

Western civilizations on the basis of technological superiority (Adas 1989) or unique quantitative

perceptions of reality (for a critique of this view, see Hart 2000). Recent cultural studies of

science have accordingly developed the concept of “translation” to dismantle monolithic notions

of “the West” and its “science,” fundamentally revising traditional historiography of the cultural

encounters between Western explorers, merchants, missionaries or colonial administrators and

non-Western societies (Prakash 1999; Hart 1999).

8

The research proposed here will examine late nineteenth-century astronomy as a culture,

governed both by internal rules and constraints as well as external needs to communicate with

other scientific and institutional cultures. Archivally, this research will investigate the particular

settings in which individual astronomers worked to produce articles, lectures and, importantly,

maps that recorded their observational findings regarding Mars’ geography. Analytically, it will

elucidate the intertwining of particular national, institutional, and social contexts with

astronomers’ scientific activities. For instance, the proposed research will investigate the ways in

which astronomers’ use of modern cartographic conventions may have functioned as an attempt

to shore up astronomy’s (and astronomers’) disciplinary status in the face of increasing

imperialist hype and funding for natural sciences such as geography. Analysis of the interactions

among astronomers of differing nationalities, competing institutions, and varying social groups

will focus on the localized contestation and negotiation of particular knowledge claims through

both texts and maps. This focus will provide a critical view of the ways in which astronomers

positioned themselves and defined their scientific identity through their studies of Mars.

In addition, the proposed research will investigate the cultural interactions among Mars

astronomer-geographers and other intellectuals in related scientific and philosophical disciplines.

Applying a science studies approach, the debated acceptance of certain astronomers’ statements

regarding the existence of a canal network on Mars’ surface can be examined as a process of

translation and negotiation. Examination of the publications and direct communications between

individuals who interpreted the Martian canals as evidence of aliens and those who subscribed to

a metaphysical belief that humans were alone in the universe will help determine whether these

groups engaged in strategies of subversive appropriation and modification of each other’s claims.

If so, the textual and cartographic record of how such claims were translated and negotiated will

be probed for evidence of the extent to which the discourse regarding Mars’ canals produced new

cultural worldviews.

9

Finally, this research will investigate the particular characteristics of the negotiated view

of Mars as a “plural world,” inhabited by humanoid “Others.” Although the late nineteenth-

century discourse regarding Mars’ inhabitants clearly employed notions of difference, familiarity

and superiority – elements that Said (Said 1978) identified as central to the modern Western

project of knowledge production – numerous astronomers and their allies formulated these

concepts differently, postulating that Martians were actually superior to humans. In this sense, I

argue, nineteenth-century Mars astronomy may have constituted an alternate modernity, one that

in fact interacted significantly with the contemporaneous imperialist modernity. Using a cultural

studies approach, this negotiation of modernities will be investigated as a process of cultural

translation, discernible in the historical record through publications by and communications

among representatives of the various modernities.

Research Plan and Methodology

Using methods of historical and archival research, this inquiry will be carried out by

examining three specific conflicts in the representation of Mars: (1) the controversy over Mars’

nomenclature, which focuses mainly on the contentious transition from Richard Proctor’s 1867

surname-based scheme to Giovanni Schiaparelli’s 1877 classical Latin convention based on the

geography of the ancient Mediterranean world; (2) the vigorous debate over the existence of

Martian canals between Schiaparelli and Nathaniel Green, whose 1877 maps differed widely in

level of detail; and (3) the conflicts regarding sensationalism during the “Canal Craze” of the

1890s and early 1900s, consisting mainly of attacks on Percival Lowell by skeptical astronomers

and other scientists who refuted his interpretations of the Martian canals as the work of intelligent

beings. Each of these controversies marked a turning point in Mars astronomy (Sheehan 1996)

and thus represents a rich opportunity for detailed analysis of the research questions outlined

above.

10

Archival Research

To analyze these controversies, I will travel to relevant libraries and observatories (see

below) to view astronomers’ original maps, sketches, manuscripts, observation logbooks,

correspondence, lectures and other materials in order to reconstruct the specific historical and

social contexts that influenced their work. In addition, I will compare representations of Martian

geography with the numerous geographical discourses presented in secondary sources that have

been interpreted from contemporaneous maps and documents, such as those produced for

imperial purposes. Finally, I will review original sources at each repository (and related archives,

where necessary) for evidence of the extent to which discourses of Mars’ geography infiltrated

other scientific and popular dialogues during the same time period. This will include examination

of newspaper articles, popular essays, books and other materials.

Repository/Location Collections

Lowell Observatory LibraryFlagstaff, Arizona

Percival Lowell collection (1894-1916)• Observation log books• Correspondence• Published manuscripts• Lectures• “Mars craze” clipping file

Royal AstronomicalSociety LibraryLondon, UK

RAS Letters• Richard Proctor correspondence• Other Mars-related correspondenceRAS MSS• Nathaniel Green’s Mars maps/drawings

1877-1888 and assorted personal papersRAS Papers• Proctor’s / Green’s publications• Mars maps, various astronomers• Scientific and popular journals• RAS lectures• Mars globes• Cartoons

Brera Observatory ArchiveMilan, Italy

Giovanni Schiaparelli collection• Observation logbooks• Published manuscripts• Mars drawings/maps• Correspondence

11

Interpretation and Analysis

Interpretive analysis of these archival materials will remain open to various cultural and

scientific explanations for the nature and meaning of what appear today to be rather curious and

remarkable geographies of the Martian landscape. The interpretive focus, however, will

primarily investigate the possibility that nineteenth-century Mars astronomy may have

significantly challenged dominant discourses of modern Western superiority by representing the

planet as a landscape inhabited by beings with superior engineering and organizational skills.

In order to assess the validity of this preliminary hypothesis, Said’s methodological

approach to discourse analysis will be used to identify narrative voice, literary structures, figures

of speech, images, themes, and motifs evident in individual scientific texts. Although Said did

not claim that his methods of discourse analysis could be applied to maps or other images, this

research will methodologically follow the examples of scholarship in the history of cartography

that suggests the identification and interpretation of conventions such as scale, framing, selection

and coding (Harley 1988; Cosgrove 1999) can be used to interpret maps as texts. To determine

the ways in which tropes and conventions used in individual texts and maps constituted (or did

not constitute) a broader discourse, Mars representations will be examined in relation to one

another. As Said showed for the Orientalist literary canon, I expect to be able to demonstrate that

certain representational conventions became broadly established in the Mars-related scientific

literature over time, especially when they metaphorically presented Mars’ geography in familiar

(terrestrial) terms.

Accepting Said’s (1978) premise (drawn from Foucault) that knowledge reflects and

maintains power, this project will seek to interpret the ways in which nineteenth-century

representations of Mars, which had widespread scientific and cultural significance at the time

they were created, influenced the hegemonic power of the modern Western nations in which such

representations were produced and consumed. In this regard, analysis of the Mars discourse will

be situated alongside analyses of contemporaneous orientalist discourses, such as Said’s argument

12

that imaginative geography produced the West’s superiority complex, Mitchell’s (1989) argument

that Western forms of representation themselves constitute a “method of order and truth”(236),

and even Bhabha’s (1995) hypothesis that European modernity was fashioned by its encounter

with the colonial Other.

In conducting an analysis of the Martian geography discourse, however, this research will

carefully avoid the pitfall of treating Western astronomy as if it were a unified endeavor.

Although I argue that a dominant discourse emerged to represent Mars, this research will

specifically focus on resistance and controversy (regarding the nomenclature applied to surface

features, for example) as a way of highlighting the heterogeneity of approaches to the

presentation of Mars. Learning from both Said’s mistakes and the helpful corrections provided

by subsequent colonial discourse analysts (Lowe 1991; Pratt 1992), this research will analyze the

ways in which competing representations of Mars were produced by astronomers working in

different professional and cultural settings, writing for different audiences. The resolution of

various controversies in favor of certain astronomers’ opinions over others’ will be assumed to

reflect a variety of power relationships that can be read through the discourse of the maps and

texts.

Preparation to Conduct the Proposed Research

In preparation for undertaking this research, I have begun communicating with

librarians/curators at repositories that hold the papers and maps of Proctor, Green, Lowell and

their contemporaries. In initial contacts, I have verified the extent and accessibility of their

collections, and have received enthusiastic support for my dissertation research. Although

contact has not yet been established with the Brera Observatory, where Schiaparelli’s papers and

maps are held, other researchers familiar with the facility have assured me that it will be

accessible and suitable for my research inquiries. Given this initial legwork, I propose that the

13

archival research activities outlined above can reasonably be accomplished according to the

timeline below.

Methodological Preparation

As a doctoral student, I have taken methodological coursework in historical/ archival

research and discourse/metaphor analysis that uniquely prepares me to undertake a complex

research design that will rely on historical interpretation of a number of disparate archival

sources. During this time, I have also completed two projects in original discourse analysis, both

of which resulted in well-received papers that have allowed me to clarify my understanding of

methodological nuances.

Language Competency

The majority of the documents and maps I intend to examine are in English, as Proctor,

Green and Lowell published and corresponded primarily in English. Schiaparelli’s work was

published mainly in Italian, however, and some of his correspondence (especially with English

and American colleagues) is in French. Accordingly, I have begun study of Italian this semester,

with a focus on reading skills, intending to attain the equivalent of second-year Italian by the time

I visit the Italian archives. My working knowledge of both Spanish and Portuguese will help me

swiftly achieve reading knowledge of Italian before summer 2004. To analyze correspondence

written in French, I intend to rely on existing or commissioned translations.

Timeline for Research Activities

Phase 1: January – April 2003 (Austin, Texas)• Proposal defense, advancement to doctoral candidacy• Secondary source readings on Percival Lowell and contemporaries• Preparations for archival research at Lowell• Italian language course• Grantwriting

Phase 2: May – June 2003 (Flagstaff, Arizona)• Archival research at Lowell Observatory

Phase 3: July – December 2003 (Austin, Texas)

14

• Analysis of research findings• Followup travel to relevant U.S. repositories, as needed• Secondary source readings on Richard Proctor, Nathaniel Green, Giovanni Schiaparelli• Preparations for archival research at RAS, Brera• Italian language courses (summer B and fall semester)• Grantwriting

Phase 4: January – May 2004 (London, UK)• Archival research at Royal Astronomical Society and related English repositories

Phase 5: June – August 2004 (Milan, Italy)• Archival research at Brera Observatory and related Italian repositories

Phase 6: September 2004 – August 2005 (Austin, Texas)• Analysis of research findings• Dissertation writing• Preparation of articles and conference presentations

Potential Funding Sources

To fund the archival phases of the proposed research, I applied in Fall 2002 or will applyin Spring 2003 for a number of grants, fellowships, and awards, including:

• Council on Library and Information Resources: Mellon Fellowship for Dissertation Researchin Original Sources ($20,000 – 12 months)

• NASA-American Historical Association: Fellowship in Aerospace History($20,000 – 12 months)

• National Science Foundation: Doctoral Dissertation Research Improvement Grant in Scienceand Technology Studies ($12,000 – 9 months)

• University of Texas Department of Geography: Teaching Assistantship($11,900 – 9 months)

• Society of Women Geographers: Evelyn L. Pruitt National Fellowships for DissertationResearch ($15,000 – 12 months)

• Royal Astronomical Society: Grants for Studies in Astronomy and Geophysics (£5,000)

• J.B. Harley Research Fellowships in the History of Cartography (£1,000 – 4 weeks)

In addition, I intend to apply in Fall 2003 for awards that would fund dissertation writing after thearchival phases of the research are complete, including:

• American Association of University Women: American Fellowships for DissertationResearch ($20,000 – 12 months)

• University of Texas: Harrington Dissertation Fellowship ($25,000 – 12 months)

15

Works Cited

Adas, Michael. 1989. Machines as the measure of man: science, technology, and ideologies ofWestern dominance. Ithaca and London: Cornell University Press.

Anderson, Benedict. 1991. Imagined communities. 2nd ed. New York: Verso.

Bhabha, Homi K. 1995. Cultural diversity and cultural differences. The post-colonial studiesreader. Editors Bill Ashcroft, Gareth Griffiths, and Helen Tiffin, 206-9. London:Routledge.

Biagioli, Mario. 1993. Galileo, courtier: the practice of science in the culture of absolutism.Chicago: University of Chicago Press.

Blaut, J. M. 1993. The colonizer's model of the world: geographical diffusionism and Eurocentrichistory. New York : Guilford Press.

Blunck, Jurgen. 1977. Mars and its satellites: a detailed commentary on the nomenclature.Hicksville, New York: Exposition Press.

Boelhower, William. 1988. Inventing America: a model of cartographic semiosis. Word andImage 4, no. 2: 475-97.

Callon, Michel. 1986 [1999]. Some elements of a sociology of translation: domestication of thescallops and the fishermen of St. Brieuc Bay. The science studies reader. Editor MarioBiagioli, 67-83. New York: Routledge.

Carter, Paul. 1999. Dark with excess of bright: mapping the coastlines of knowledge. Mappings.Editor Denis Cosgrove, 125-47. London: Reaktion Books.

Collins, H. M. 1974 [1999]. The TEA set: tacit knowledge and scientific networks. The sciencestudies reader. Editor Mario Biagioli, 95-109. New York: Routledge.

Cosgrove, Denis. 1999. Introduction: mapping meaning. Mappings. Editor Denis Cosgrove, 1-23. London: Reaktion Books.

Crowe, Michael J. 1986. The extraterrestrial life debate 1750-1900: the idea of a plurality ofworlds from Kant to Lowell. Cambridge: Cambridge University Press.

Driver, Felix. 1992. Geography's empire: histories of geographical knowledge. Environment andPlanning D: Society and Space 10: 23-40.

Edgerton, Samuel Y. 1987. From mental matrix to mappamundi to Christian empire: the heritageof ptolemaic cartography in the Renaissance. Art and Cartography. Editor DavidWoodward, 10-50. Chicago: University of Chicago Press.

Edney, Matthew H. 1997. Mapping an empire: the geographical construction of British India,1765-1843. Chicago and London: The University of Chicago Press.

16

Godlewska, Anne. 1995. Map, text and image: The mentality of enlightened conquerors: a newlook at the Description de l'Egypte. Transactions of the Institute of British Geographers20: 5-28.

Grove, A. T., and Oliver Rackham. 2001. The nature of Mediterranean Europe: an ecologicalhistory. New Haven: Yale University Press.

Grove, Richard H. 1997. Ecology, climate and empire. White Horse Press.

Guthke, Karl S. 1983. The last frontier: imagining other worlds, from the Copernican revolutionto modern science fiction. Translator Helen Atkins. Ithaca and London: CornellUniversity Press.

Harley, J. B. 1988. Maps, knowledge, and power. The iconography of landscape: essays on thesymbolic representation, design and use of past environments. Editors Denis Cosgrove,and Stephen Daniels, 277-312. Cambridge: Cambridge University Press.

———. 1989. Deconstructing the map. Cartographica 26: 1-20.

Hart, Roger. 1999. Translating the untranslatable: from copula to incommensurable worlds.Tokens of exchange: the problem of translation in global circulations. Editor Lydia H.Liu, 45-73. Durham and London: Duke University Press.

———. 2000. The great explanadum, review of Alfred W. Crosby's The Measure of Reality:Quantification and Western Society 1250-1600. American Historical Review 105, no. 2:486-93.

Helgerson, Richard. 1988. The land speaks: cartography, chorography, and subversion inRenaissance England. Representing the English Renaissance. Editor Stephen Greenblatt,327-61. Berkeley: University of California Press.

Hudson, Brian. 1977. The new geography and the new imperialism: 1870-1918. Antipode 9, no.2: 12-19.

Kuhn, Thomas S. 1970. The structure of scientific revolutions. 2nd ed., Vol. 2. InternationalEncyclopedia of Unified Science, eds. Otto Neurath, Rudolf Carnap, and Charles Morris,2. Chicago: University of Chicago Press.

Lowe, Lisa. 1991. Critical terrains: French and British orientalisms. Ithaca, NY : CornellUniversity Press.

Manners, Ian. 1997. Constructing the image of a city: the representation of Constantinople inChristopher Buondelmonti's Liber Insularum Archipelagi. Annals of the Association ofAmerican Geographers 87, no. 1: 72-102.

Mitchell, Timothy. 1989. The world as exhibition. Comparative Studies in Society and History31: 217-36.

Morton, Oliver. 2002. Mapping Mars: science, imagination and the birth of a world. London:Fourth Estate.

17

Prakash, Gyan. 1999. Another reason: science and the imagination of modern India. Princeton,New Jersey: Princeton University Press.

Pratt, Mary Louise. 1992. Imperial eyes: travel writing and transculturation. London: Routledge.

Ryan, Simon. 1994. Inscribing the emptiness: cartography, exploration and the construction ofAustralia. De-scribing empire: post-colonialism and textuality. Editors Chris Tiffin, andAlan Lawson, 115-30. London: Routledge.

———. 1996. The cartographic eye: how explorers saw Australia. Cambridge: CambridgeUniversity Press.

Saberwal, Vasant K. 1997. Science and the desiccationist discourse of the 20th century.Environment and History 3: 309-43.

Said, Edward W. 1978. Orientalism. New York: Pantheon Books.

Shapin, Steven, and Simon Schaffer. 1985. Leviathan and the air-pump: Hobbes, Boyle and theexperimental life. Princeton: Princeton University Press.

Sheehan, William. 1996. The planet Mars: a history of observation and discovery. Tucson: TheUniversity of Arizona Press.

JSC MARS-1: MARTIAN REGOLITH SIMULANTCarlton C. Allen1, Richard V. Morris2, David J. Lindstrom2, Marilyn M. Lindstrom, and John P. Lockwood3

1Lockheed Martin Engineering & Sciences, Houston, TX 77058 2NASA Johnson Space Center, Houston,TX 77058 3Hawaiian Volcano Observatory, Hawaii Volcanoes NP, HI 96718

We have developed a simulant to the regolith of Mars for support of scientific research, engineeringstudies, and education. JSC Mars-1 is the <1 mm size fraction of a palagonitic tephra (glassy volcanicash altered at low temperatures). The material was collected from the Pu’u Nene cinder cone, located inthe saddle between Mauna Loa and Mauna Kea volcanoes on the Island of Hawaii. Palagonitic tephrafrom this cone has been repeatedly cited as a close spectral analog to the bright regions of Mars [1,2,3].

Simulant Preparation and Analysis. The tephra was mined from a cinder quarry on the slope ofPu’u Nene cone. Soil overburden was removed and tephra was collected from a palagonitized zone 40-60 cm thick. The tephra was dried and sieved to separate the <1 mm size fraction. This material waspackaged in moisture-proof containers for shipping and storage.

Preliminary Simulant Characterization. We analyzed a single sample from the area chosen forlarge scale simulant preparation. Splits were characterized by visible and near-IR (VIS/NIR) reflectancespectroscopy at the Johnson Space Center. X ray fluorescence (XRF) and loss on ignition (LOI)analyses were performed at Washington State University. We intend to publish detailed data fromrepresentative samples of JSC Mars-1 in the near future.

Spectra. JSC Mars-1 is yellow-brown in color. Figure 1 compares the VIS/NIR spectrum of thesimulant to a composite martian bright region spectrum (atmospheric contributions removed) [4]. Bothspectra contain a relatively featureless ferric absorption edge through the visible, an indication of a ferricabsorption band in the 800-900 region, and relatively flat absorption in the near-IR. Bands at 1400 and1900 nm in the simulant spectrum result from higher levels of H2O and OH in the simulant than on Mars.The presence of the ferric features near 600, 750 and 860 nm in the martian spectrum imply higherlevels of red (well-crystalline and pigmentary) hematite on Mars than in the simulant [5,6].

Chemical Composition. Table 1 lists the major and minor oxide composition of JSC Mars-1, asmeasured by XRF. This composition is compared to that of a typical Mars surface sample analyzed atthe Viking lander 1 (VL-1) site [7].

Mineralogy. Morris et al. [3] published extensive analyses of a <1 mm tephra sample collected fromPu’u Nene. The sample is dominated by amorphous palagonite. The only phases detected by X raydiffraction are plagioclase feldspar and minor magnetite. These analyses constrained the abundance ofphyllosilicates to <1 wt.%. Iron Mossbauer spectroscopy detected magnetite as well as traces ofhematite, olivine, pyroxene and/or glass. The majority of iron was present as nano-phase ferric oxide(64%). These data yield a Fe2+/Fe3+ ratio of 1/3.

Grain Size. Table 2 lists the published grain size distribution of Pu’u Nene tephra [3]. Forcomparison, the blocky material which covers 78% of the area near VL-1 on Mars ranges in size from0.1-1500 m [8].

Specific Gravity. The bulk specific gravity of JSC Mars-1 is 0.8 g/cm3. This value can be increasedto 0.9 g/cm3 by vibrating the sample. The drift material near VL-1 has a specific gravity of 1.2 +/- 0.2g/cm3 and the blocky material has a value of 1.6 +/- 0.4 g/cm3 [8].

Magnetic Properties. JSC Mars-1 contains a highly magnetic component. Approximately 25 wt.% ofthe sample can be lifted with a strong magnet. By comparison, observations of the Viking sample armmagnets indicate that the martian soil contains between 1-7% magnetic material [9].

Availability. We anticipate that approximately 9,100 kg (20,000 lb) of JSC Mars-1 will be available in1997 for distribution to qualified investigators and teachers. The simulant will be stored at the JohnsonSpace Center. Anyone desiring a portion of this material should address their request to Dr. CarltonAllen (address above; telephone 281-483-2630, fax 281-483-5347).

References. [1] Evans, D. L. and Adams, J. B. (1979) Proc. 10th Lunar Planet. Sci. Conf. 1829-1834.[2] Singer, R. B. (1982) J. Geophys. Res. 87, 10,159-10,168. [3] Morris, R. V. et al. (1993) Geochim.Cosmochim. Acta 57, 4597-4609. [4] Mustard, J. F. and Bell, J. F., III (1994) Geophys. Res. Lett. 21,3353-3356. [5] Morris, R. V. and Lauer, H. V., Jr. (1990) J. Geophys. Res. 95, 5101-5109. [6] Morris, R.V. et al. (1997) J. Geophys. Res. in press. [7] Clark, B. C. et al. (1982) J. Geophys. Res. 87, 10,059-

Lunar and Planetary Science XXVIII 1797.PDF

JSC MARS-1: MARTIAN REGOLITH SIMULANT: Allen, C.C. et al.

10,067. [8] Moore, H. J. et al., (1987) U.S.G.S. Prof. Paper 1389. [9] Hargraves, R. B. et al., (1977) J.Geophys. Res. 82, 4547-4558.

Figure 1. VIS/NIR reflectivity spectra of Mars Composite Bright Region [4] and JSC Mars-1

Table 1. Chemical Composition

MartianJSC Mars-1 Surface

Fines C-1

Oxide Wt.%* Wt.%** Wt.%***SiO2 34.5 43.7 43Al2O3 18.5 23.4 7.5TiO2 3.0 3.8 0.65FeO 2.8 3.5 n.d.Fe2O3 9.3 11.8 17.6MnO 0.2 0.3 n.a.CaO 4.9 6.2 6MgO 2.7 3.4 6K2O 0.5 0.6 0Na2O 1.9 2.4 n.a.P2O5 0.7 0.9 n.a.SO3 n.a. n.a. 7Cl n.a. n.a. 0.7

LOI**** 21.8

Table 1. (continued)

n.d. not detected n.a. not analyzedFe2+/Fe3+ = 1/3 * XRF ** XRF normalizedwithout LOI *** Ref [7] **** Weight loss afterheating for 2 hrs in air at 900 C (includes H2O,SO3 , Cl)

Table 2. Pu’u Nene Tephra Grain Size*

Size ( m) Wt.%

500-1000 21.4250-500 29.5150-250 20.890-150 12.945-90 9.220-45 5.4<20 1.3

* Ref [3]

Lunar and Planetary Science XXVIII 1797.PDF

Mars Exploration "Follow the Water"

Young Ho Park

Jet Propulsion Laboratory Pasadena, CA 91109

Abstract— The red planet Mars has been a subject of imagination over the centuries, as well as intense scientific interest. As the overwhelming success of two Mars Exploration Rovers unfold before us, this article reviews the overview of NASA's Mars Exploration and rationale. 1. INTRODUCTION In 2004, we have observed two historic events in Mars exploration. The first Mars Exploration Rover (named Spirit) landed on Mars on January 3, 2004. The second Mars Exploration Rover (named Opportunity) landed on Mars on January 24, 2004. At the time of this writing, both rovers are operating nicely, taking pictures of Mars surface and taking various scientific measurements to reveal many secrets of Mars now and many, many years ago. There have been only five successful landing of spacecraft on Mars surface: Viking 1 Lander 1 and 2 in 1975, Mars Pathfinder with Sojourner in 1997 and Spirit and Opportunity this year. As we know that many Mars missions have failed, Mars missions are challenging and require extreme ingenuity and dedication of all involved team. 2. MARS VS EARTH Mars is the fourth planet from Sun. The distance from Sun is about 1.5 times that of Earth. The mass of Mars is 10% of Earth. The diameter of Mars is 53% of Earth. The gravity of Mars is 37% of Earth. The Mars atmospheric pressure is only 0.7% of Earth atmosphere. The average recorded temperature on Mars is -63° C with a maximum temperature of 20° C and a minimum of -140° C. The atmosphere of Mars is quite different from that of Earth. It is composed primarily of carbon dioxide with small amounts of other gases. The six most common components of the atmosphere are: Carbon Dioxide (95.3%), Nitrogen (2.7%), Argon (1.6%), Oxygen (0.13%), Water (0.03%), Neon (0.00025 %).

Figure 1. Earth and Mars 3. WHY MARS? Mars is the only planet, other than Earth, that shows strong evidence of liquid water having coursed over its surface. There are many clear signs of rivers and lakes on Mars surface (Figure 2). Based on limited Mars exploration, it seems that there is no obvious sign of water on Mars surface at this time. However, there is abundance of indication that once water flowed on Mars surface at one time or another in the long history of Mars. Although our current understanding of life's origins may be limited, at least on Earth, there is life where water is. Thus, based on we see on Mars surface it is possible that Mars may have been habitable and may have harbored life.

In Figure 3, striking features of gullies are shown in the picture recently taken by Mars Orbiter Camera on Mars Global Surveyor. According to a dictionary, a gully is "a deep ditch or channel cut in the earth by running water after a prolonged downpour". No one is sure yet how the gullies are formed. One conjecture is that subsurface water or ice melted and the water may have gushed out. The implication of this conjecture is tremendous so that there may be water or ice under the Martian surface even at present time. An analogy may be permafrost on Earth in polar region such as Alaska. The permafrost is permanently frozen subsurface soil. However, caution may be required to draw such haste conclusion without further conclusive evidence.

NASA has created Mars Program with a theme of " Follow the Water." The objective of the program is to detect conclusive evidence whether water existed on Mars, water exists on Mars subsurface if not on surface, and ultimately evidence whether life, even in microbiological life form, existed in the lifetime of Mars.

Figure 2. Mars Surface picture taken by a Mars Orbiter

Figure 3. Picture taken by Mars Orbiter Camera on Mars Global Surveyor 4. MARS EXPLORATION ROVER Mars Explorer Rover (MER) mission is to send two rovers to Mars. The scientific objective is to determine the history of climate and water at sites on Mars where conditions may once have been favorable to life. Each rover is equipped with a suite of science instruments that will be used to read the geological record at each site, to investigate what role water played there, and to determine how suitable the conditions would have been for life.

Figure 4. Artist Conception of a Mars Exploration Rover. Mars Exploration Rover A (now called Spirit) was launched on June 10, 2003 and landed on Mars on January 3, 2004. Mars Exploration Rover B (now called Opportunity) was launched on July 7, 2003 and landed on Mars on January 24, 2004. Both rovers are identical in design. The names of rovers were suggested by a schoolgirl and selected after a worldwide competition. As seen in media, both rovers are conducting their scientific mission among many challenges. Each rover has 90 Martian days for it's prime mission. Scientific instruments of each rovers are: Panoramic Camera, Mini-Thermal Emission Spectrometer, Microscopic Imager, Moessbauer Spectrometer, Alpha Particle X-Ray Spectrometer. The robotic arm includes rock abrasion tool. Also, each rover has magnet arrays. (For more information visit http://marsrovers.jpl.nasa.gov/)

MER Spirit has landed in Gusev crater area (Figure 5) and the current Mars surface does not seem to have liquid water. However, the scientists believe that Gusev crater area included flowing water, accumulated water in lakes, and deposit of sediment over a long period of time. This history makes Gusev crater very interesting exploration site. Figure 6 shows the picture where MER Spirit examines a Mars Rock.

MER Opportunity has landed in a small crater in Meridiani Planum (Figure 5). Meridiani Planum interests scientists because it contains an ancient layer of hematite, an iron oxide that, on Earth, almost always forms in an environment containing liquid water. The site appears dry now. So how did the hematite get there? Was there once water in the area? If so, where did it go? These are main questions for which MER Opportunity will collect in-situ measurement data.

5. FUTURE AND SUMMARY In addition to two Mars rovers on Martian surface, NASA has two spacecraft orbiting Mars now: Mars Global Surveyor and Mars Odyssey. Also, European Space Agency has Mars Express in Mars Orbit currently. NASA plans to send one spacecraft to Mars every two years. Phoenix (a lander) will be launched in 2007 and Mars Science Laboratory (MSL, a rover) will be launched in 2009. In a long term, a Mars sample return mission is considered.

Figure 5. Landing Site for Mars Exploration Rovers

Figure 7. Layered Rock picture taken by MER Opportunity Figure 6. MER Spirit examining a Mars Rock

Of course, President has set a long-term goal for sending men to Mars possibly within two decades. People ask, " Why do we do this for such a high cost?" Practical answer is that there are invaluable science and technological byproduct. Teaching science and technology and inspiring next generation are another essential part. On the other hand, continuous advancement of a civilization is only possible with the spirit of "Exploration".

14307

Radar Ranging of the Planet Mars at 8495 MHzG. S. Downs and P. E. Reichley

Communications Systems Research Section

A simulation was performed for the radar system in order to ensure detectionof the planet Mars at the start of the 1975 series of radar probes of the surface.Appropriate parameters were found. Appropriate parameters were also found foruse at opposition (December 1975). Systematic errors in the measured delay withchanges in surface roughness were observed. This effect is shown to be manytimes larger than the expected rms fluctuations in the measured delays.

\. Introduction

The surface of the planet Mars will be probed withS- and X-band radar signals during 1975 and 1976, all inpreparation for the landing of two spacecraft in 1976.The measurements to be performed at JPL will use theR & D radar system at DSS 14, Goldstone, California, ata frequency of 8495 MHz. In preparation, several seriesof radar data sets were simulated and processed. That is,the expected power in the reflected signal was calculatedas a function of time delay and frequency for several setsof radar system parameters. The random fluctuationscaused by receiver noise were superimposed using arandom number generator. The appropriate signal-to-noise ratio was calculated for certain distances to theplanet, assuming appropriate X-band antenna param-eters. The round-trip time delay of the signal reflected

from the surface of the planet was determined. Up to 120delay measurements were made to allow determinationof the detection probability and the measurement accu-racy. An appropriate set of radar parameters was chosenin light of the results of this study.

II. How Delay Is Measured

The basic radar data set consists of a matrix of re-ceived power as a function of time delay and excess dop-pler shift. Separate regions on the surface will exhibitseparate time delays because of the spherical shape ofthe planet.

Let 9 represent the angle between the line of sight fromthe radar to the center of the planet and the radius to the

JPL DEEP SPACE NETWORK PROGRESS REPORT 42-29 95

point on the surface. The time delay ^ of the signal re-flected from that surface point back to the radar is givenby

(1)

where R is the radius of the planet and c is the speed oflight; D is the distance along the line of sight to theclosest point on the surface. Note that 6 is also the angleof incidence of the signal on the surface. 6 and D areeach functions of time. The large delay due to D is re-moved at the time of data collection. The small changesin 6 due to the relative motion of Mars and Earth in thetime interval 2D/c are neglected. The locus of a curveof a constant r is a circle in the plane perpendicular tothe line of sight and whose center lies on the line ofsight. Regions of a particular delay are isolated by modu-lating the phase of the transmitted signal with a pseudo-random binary code. If the bit length of the code is Tseconds, the normalized cross correlation function of thetransmitted signal with the received signal is given by

RrM '•

- -^ IT - T« I; -T0| <T

- T O ! >T

(2)

where TO is the round trip delay. A band of surface ele-ments located within T seconds of TO can be isolated. Inpractice, the received signal is passed through a bank ofcorrelators, the tth corresponding to a round trip delayT0 j . The power in the output of the ith correlator is thenthe sum of the power in the reflected signals correspond-ing to the delays r0i — T <T < roi + T, each delay com-ponent being weighted by the range window R(T) —Rf.(r). There is an ambiguity in the delay TO of the re-flected signal which is equal to the length of the binarycode. That is, R(T) is periodic in time mT, where m is thenumber of bits in the code. Usually the planetary ephem-eris is good enough to resolve this ambiguity in time de-lay. There remains, however, a small perturbation to thetotal reflected power from the region defined by R(T,,,)since contributions are obtained from all regions centeredat T = TO, ± nmT, where n is any integer.

Further resolution of the planet's surface is usually ob-tained by taking advantage of the doppler spreading ofthe reflected signal caused by the rotation of Mars. Lociof constant doppler shift are circles on the surface, par-allel to the plane defined by the apparent spin axis of theplanet and the line of sight. The intersections of the

circles of constant range and the circles of constant dop-pler shift then isolate particular small regions on the sur-face (see Fig. 1). Therefore, the discrete power spectrumof the output of each correlator is estimated using dis-crete time samples. The power at each discrete frequencyfa in the spectrum is in fact the sum of power from regionsof varying doppler shift /. Each region is weighted by thefamiliar function

(3)

where N is the number of discrete frequencies, separatedby A/, in the power spectrum.

The delay-doppler geometry imposed on the planet'ssurface is pictured in Fig. 1, where the planet is viewedfrom the direction of the apparent spin axis. The angleof incidence 6 is the angle between the direction to Earthand IV, the local normal to the mean sphere. The shapeof the range window R(T) is shown at the left for the caseof T = 6 /iS. The shape of the doppler window D(/) isshown at the right for the case of A/ = 10.2 Hz, whenA. = 12.55 cm. As is often the case in these measurements,the correlators are separated such that TOU + D — roi = T/2.Note that as the planet rotates, any given region near thedoppler equator (a great circle perpendicular to the cir-cles of constant doppler shift and bisecting the rings ofconstant range) is probed at a variety of angles of inci-dence. Each set of received power as a function of delayand doppler (a data frame) represents a snapshot of thesurface near the doppler equator. An example of such adata frame appears in Fig. 2.

Consider for the moment the power vs time delay at afixed doppler shift. The reflected power corresponds to anarrow region at a particular longitude and a minimumangle of incidence. The location of the planet at that par-ticular doppler shift is taken to be the delay at whichthe power is a maximum. As the planet rotates, a seriesof delay functions corresponding to particular values of6 become available for a particular longitude. With theextra delay due to curvature of the surface removed, thecollection of delay functions is usually added to producea composite delay function in which the signal-to-noiseratio is larger. The delay corresponding to the peakpower then defines the distance to the planet.

In practice, the presence of noise will cause errors inlocating the planet. Also, the weak signal case gives riseto a finite possibility of a false detection of the planet.

96 JPL DEEP SPACE NETWORK PROGRESS REPORT 42-29

The following discussion is devoted to a determinationof the probability of a correct detection and the measure-ment error, once detection has been established. Theseprobabilities and errors are dependent on the signal-to-noise ratio, which in turn is dependent on the parametersof the radar system and the surface characteristics. Theradar system parameters and surface characteristics arevaried over an appropriate range in the discussion below.

III. The Data

An example of simulated data is presented in Fig. 2.The spectrum of the output of 32 correlators is presentedin Fig. 2(a) for the noise-free case. The magnitudes ofthe spectral components are proportional to the receivedpower expected for the case of T — 6 /*s and A/ = 36.2 Hzat A = 3.53 cm. Note that correlators are offset progres-sively in delay by T/2 = 3 ps. These magnitudes werecalculated by evaluating the radar equation:

P,GJ.GrA2 p0C

// ds. R(r - r0) D(f -

(4)

where P, and G, are the transmitting power and antennagain, respectively. G, is the receiving antenna gain. Thetransmitter operates at wavelength A. The planet, locatedat distance D, is characterized by a reflectivity p0 androughness parameter C. The denominator of the inte-grand in Eq. (4) (sometimes called the Hagfors back-scatter function, Refs. 1 and 2) describes how the powerscattered back toward the receiver varies with the angleof incidence, where 0 is now a function ^ and /. The sur-face S over which the integration takes place is deter-mined by the position of the range and doppler windowsrelative to the planet. The details of the evaluation of theintegral, and in particular the effects of aliasing in fre-quency and ambiguity in range will be discussed in aforthcoming article.

The received power P(f,r) was calculated for 32 de-lays and 64 frequencies for values of A/ = 36.2, 72.5 and145 Hz with T = 6 us and for A/ = 1'45 Hz with T = 12^s. In each case, T O ( ^ I I — T0i = AT — T/2. In the case ofMars, these values of A/ correspond to N — S slices on thesurface of widths of 0.16, 0.32 and 0.64 deg in longitude,respectively. The transmitter power Pz was taken to be400 kW, operating at 8495 MHz, so A = 3.53 cm. Thegains GT and G, were taken to be equal, and these valueswere deduced from data provided by Freiley (Ref. 3).

A nominal value of system efficiency of 0.40 was assumedfrom the data of Ref. 3. This value corresponds to a value1 a lower than the mean transmit system efficiency at anelevation of 70 deg. The corresponding antenna gains are71.1 dB. The 4 sets of values of (A/,T) were subdividedinto sets of different values of D, where D varied between0.56 AU (closest approach during the 1975 opposition)and 1.5 AU. These subsets were, in turn, subdivided evenfurther to correspond to roughness C = 50, 150, 300, 1000,2000 and 5000. A total of 78 distinct sets of data frameswere then generated from Eq. (4). A value of pa = 0.08was assumed for all evaluations of Eq. (4). This value isan average obtained from previous radar probes of Mars.Although p0 varies between 0.01 and 0.15, most regionshave a reflectivity close to 0.08.

In practice, data frames similar to those of Fig. 2(b)are measured at regular time intervals. To improve thesignal-to-noise ratio, several sequential data frames areusually added together. The magnitude of each spectralcomponent then represents an energy. That is, a signal ofintensity P — P(f,r) watts is integrated for t, seconds toproduce Ft, joules. Now, t, should be long enough to al-low a particular region on the planet's surface to rotatefrom one discrete doppler frequency to the next. In thisanalysis, t, was chosen such that the planet rotated about0.75 of that distance. For example, when A/ = 36.2 Hz(0.16 deg in longitude), t, was chosen to be 30 seconds,an interval in which the planet rotates 0.12 deg.

The superimposing of a noisy signal of the propermagnitude was performed in the following manner. Aseries of random numbers with a variance of 1 .0 was gen-erated. The scale of the variance was chosen by notingthat the receiver noise power is fcT8Af watts for eachspectral component, where k is Botzmann's constant andTs is the system noise temperature. In time t, the meanenergy obtained by integrating this component of thenoise is kT,±ft, and the variance associated with themeasurement of this component is (kT,)2£ft,. The mea-surement of the planetary component Pt, also is subjectto random fluctuations. The total variance is calculatedassuming that the wideband receiver component and theplanetary component are each nearly Gaussian randomprocesses. (They are in fact Rayleigh processes in whichthe mean is much larger than the root-mean-square fluc-tuation.) The total variance is then (kT,Ytft, + (Pt,)2/

In the generation of the series of random numbers itwas assumed that Pt, < < kT&ft,, so that the planetarycomponent of the variance could be ignored. The series


of random numbers, so scaled, and the mean value of thereceiver noise were added to the function P(/,T) to obtaina simulated data frame. T, was taken to be 23 K. A dataframe similar to Fig. 2(b) was generated from the noise-free frame of Fig. 2(a) in the above manner. The data inFig. 2(b) correspond to D = 0.56 AU and C = 300. Theconstant wide-band component of the receiver noise hasbeen subtracted. A total of 70 data frames containingindependent additive noise, representing integrations overti seconds, were generated to provide a good measure ofthe statistics of interest.

IV. The Delay Measurements

In practice, one cannot probe the planet's surface witha monostatic radar system (one antenna) on a continuousbasis, since the transmitter must be turned off duringreception. Hence, the simulated data frames were ar-ranged in time to duplicate the case in which the radarsignal is transmitted for a time interval equal to theround-trip time between Earth and Mars, and then re-ceived for one round-trip time. Reception immediatelyfollows transmission for an equal time interval. Clearly,in one round-trip time one will usually collect severaldata frames, each representing an integration of t, sec-onds. Since the transmitter is not on continuously, only% of the available angles of incidence will be probed. Aview window, of width equal to the amount of rotationaccomplished in one round-trip time, slides over the sur-face allowing some angles of incidence and omittingothers.

The results of the delay measurements are presentedin Tables 1-4. At first, the radar system parameters T andAf were chosen to be equal to those used in earlier workon Mars at 2388 MHz. The doppler shifts were scaledfrom 2388 to 8495 MHz such that A/ corresponded to alongitude interval AL of 0.16 deg. At each of the fourvalues of D in Table 1, six values of roughness C werechosen to cover the range expected at 8495 MHz. A num-ber NT (between 90 and 120) composite delay functions,each representing a different mixture of angles of inci-dence, were available for each combination of D and C.If the peak amplitude of a composite delay function was2.5 to 3 times larger than the rms noise level, the planetwas considered detected and the corresponding delay Twas recorded. However, in ND detections, a certain num-ber NF are false detections which usually occur in theweak signal case. Values of T which placed the planetmore than T fis away from the known position of the

planet were considered false detections. The fraction Pjof successful detections and the fraction Pe of these de-tections that were false are listed in Table 1 and calcu-lated from

Np-N,

NT

(5)

The values of ND, the mean TJO in the range estimates,and the associated rms fluctuation o-TO presented in Table1 represent averages over all the available data. However,some of the composite delay functions contain contri-butions from angles of incidence primarily near 0 deg(the maximum is about 2 deg in this simulation). Thesedelay functions were isolated and, finding Nc of them, themean T&C of this set and the associated rms fluctuations<TTC were calculated and listed in Table 1.

The true mean and variance are represented by Tt,cand <TTC. The reason for this can be understood by examin-ing Fig. 3. The delay T of each composite delay functionhas been presented vs the centroid nd of the angle-of-incidence view window. The centroid is expressed inunits of AL (or A/). In Fig. 3, where C = 5000, the shapeof the backscatter function changes rapidly with 6, so theshape of the delay function flattens markedly as rid in-creases. A positive drift in the delay of the peak ampli-tude and a decrease in the magnitude of the peak accom-pany this flattening. The positive drift is readily observ-able in Fig. 3(a). Noise fluctuations are low since thecorresponding distance is only 0.56 AU. The distance is1.14 AU in Fig. 3(b), and the larger scatter in T for n<; > 6is evidence of the decreasing peak amplitude. By includ-inly only the values of r for which nd < 6 in the statistics,a true representation of the measurement accuracy isobtained. Including all available values of T produces alarger variance because of the systematic drift in T withwindow position.

The mean n,c is a function of the roughness parameterC. In Fig. 4(a) the measurements of T are presented vsnd for C = 150 at a distance of 0.56 AU. The drift in rwith nd is not as extreme as in Fig. 3(a) since the back-scatter function varies more slowly with 6. Note howeverthat TJ,C for nd < 6 is not identical to the similar value inFig. 3(a). This bias is again caused by the retarded, flat-ter delay functions characteristic of low values of C or


larger values of 6. The data of Fig. 4(b), correspondingto D = 0.8 AU, show how small signal-to-noise ratiosmask the effects discussed above.

At opposition D = 0.56 AU. The results of Table 1indicate that maximum ranging accuracies of 40 to 300ns can be obtained. The systematic changes of T with nd

and C are large compared to these hypothetical accuracies.It will then be desirable to apply corrections to the esti-mates of T to obtain the minimum possible rms fluctu-ations.

The data of Tables 3-4 were obtained in a search of thedata-frame parameters which ensure detection of theplanet Mars at the start of the 1975 series of measure-ments (during August, when Mars is at 1.2 AU). Theparameters underlying Table 4 (T = 12 ps, AL = 0.64deg) provide a reasonable probability of detection ofrough as well as smooth surfaces at 1.2 AU. These are theparameters to be used at the start of the series of mea-surements. Tables 2 and 3 are useful as an aid in deter-mining what T and AL should be as Mars progressestowards opposition.

References

1. Hagfors, T., "Backscattering from an Undulating Surface with Applications toRadar Returns from the Moon;" }GR, 69, pp. 3779-3784,1964.

2. Hagfors, T., Radar Astronomy, Evans, J. V., and Hagfors, T., eds., Chapter 4,pp. 187-218, McGraw-Hill, New York, 1968.

3. Freiley, A. J., "DSS-14 XKR Cone Performance," JPL Interoffice Memo No.3331-75-001, Mar. 10,1975 (an internal document).


Table 1. Detection probability and a- for T = 6 us, AL = 0.16 deg, p,( = 0.08

5,0.56 50

150300100020005000

0.80 50150300100020005000

1.00 50150300100020005000

1.14 50150300100020005000

r*%98100100100100100

5697100100100100

36283959592

23873869188

P*%

100000

1000000

5711000

50103000

ND

118120120120120120

65113117117117117

476102117117113

23873869188

"TV T6o>

/JS US

0.44 2.60.26 2.40.31 2.30.31 2.00.38 1.70.45 1.4

1.330.650.620.390.400.54

1.801.180.720.620.550.57

1.341.520.940.450.540.50

NO

646464645656

406161615555_

4254585552_

2939505249

"re,/tS

0.300.120.080.090.040.04

1.380.420.150.110.040.04

_

0.980.570.330.320.22

—

1.440.800.380.450.35

rte>/*s

2.42.252.11.751.41.1

_

-----

_

-

---

—----

-


Table 2. Detection probability and <rr for T = 6 /is, AL = 0.32 deg, />, = 0.08

D,AU

1.00

1.14

1.50

C

50

150300

1000

2000

5000

50150

30010002000

5000

50

150

3001000

2000

5000

%

1281

82

93

9493

4

60829394

93

_

-

307683

84

P.,

_

33

0

0

0

_

63

000

_

-100

0

0

*,13

91

839495

94

461

83949594

_

-

266773

74

"TO' rbo>

1.47

1.09 2.50.81 2.5

0.61 2.00.49 1.60.74 1.5

1.151.500.81

0.610.490.74

_ _

-

1.37

0.530.630.45

"e

8564652

47

46

_

34465247

46

_

-

20413737

/is

1.181.240.680.58

0.560.54

_

1.500.680.58

0.560.54

_

-

1.28

0.590.650.51

Tic-

IIS

—2.52.5

2.01.61.2

_

--

-

-

-

_

--

-

-

-

ORIGINAL PAGE isPOOR QUALITY


Table 3. Detection probability and OT for T = 6 /is, AL = 0.64 deg, />„ = 0.08

D,AU

1.00

1.14

1.50

C

50150300100020005000

50150300100020005000

50150300100020005000

%'

4692100100100100

11789510010098

_

949788084

%'

1210000

3160001

_

-

5000

v.

4998106106106106

107490959493_

844697175

CTTO» T6o»/US JUS

1.92 3.30.93 2.71.00 2.70.91 2.20.61 1.70.67 1.4

1.601.501.000.750.640.71

_ _

1.51.620.720.470.64

»,

316570705353

64857624950

_

727443737

"re-ps

1.720.980.650.510.290.20

2.01.550.840.530.560.82

_

1.401.200.650.460.25

Tbc>

MS

3.12.92.62.01.51.1

_

-----

_

----

-


Table 4. Detection probability and <rT for T = 12 pS, AL = 0.64 deg, P, = 0.08

AU

1.00

1.14

1.50

C

50150

3001000

2000

5000

50150300

10002000

5000

50150

30010002000

5000

%

82

100100

100

100100

44

9198

10098

95

_

3068

7680

80

%

4

00

0

00

1021

0

00

_

100

100

N0

91

111

111111

111

111

49

101109

111109106

_

3069

7781 '

80

ffTO» T60»

Its its

2.62 5.41.48 5.0

1.10 4.4

0.78 3.41.12 2.8

1.48 7.0

3.40

2.001.751.28

1.440.93

_ _

3.101.84

1.08

1.081.04

*e

86103

103103

53

53

3063

707155

51

_

1942

44

38

37

Its

2.66

1.500.74

0.700.32

0.36

3.40

2.20

1.960.92

0.771.14

_

3.601.92

1.10

1.38

1.30

T%

5.4

5.0

4.43.2

2.4

1.6

_

-

-

--

-

_

--

-

-

-


3.5,1

POSITIVEDOPPLERSHIFT

EARTH

Fig. 1. Partitioning of Mars by the radar system


90

30

15

-1.12 1.16

DOPPLEK SHIFT, kHz

Fig. 2. Received power vs doppler shift and delay for -if = 36.2 Hz, T = 6^8, p0 = 0.08,and C = 300: (a) noise-free case, (b) noisy case (see text for noise parameters)

-1.12


T i 1 1 1 1

(a)

-

• •

1 1 1 1 I 1

(b)

— x -X

x x

* * * * * * * * * * *S * w X ^ X X

1 1 1 1 1 110 10

Fig. 3. Estimation of r MS the centroid n,, of the angle-of-incidence window for C = 5000:(a) 0 = 0.56 All, (b) D = 1.14 AU

(a)

»:»*»:»*, :«"**«««

(b)

x-* x

„x

10 10

Fig. 4. Estimation of - vs the centroid n,, of the angle-of-incidence window for C = 150:(a) O = 0.56 AU, (b) D = 0.8 AU


Geophysical Research Abstracts, Vol. 9, 01794, 2007SRef-ID: 1607-7962/gra/EGU2007-A-01794© European Geosciences Union 2007

The Red Soil on Mars as a proof for water andvegetation!R.PaepeGEOBOUND International, BV MUHS, The Hague, The Netherlands

([email protected] / Phone: +31-703-520510)

Red Soils on Earth are common features. Actually, what does the label “soil” standsfor? It is used by so many people most of them with no training in soil science or inthe related field of geology. And yet, everybody uses “Soil” as a simple connotation toindicate the surface, floor, territory, base/bottom, ground; or any kind of loose earth.And when it is red, it is called Red Soil. However, as a concept in soil science a RedSoil means something quite different: it then relates to a thorough soil/pedologicalweathering most often originating under tropical climatic conditions. The latter ischaracterised by specific sediment differentiation in the uppermost geological layersas to structure, texture and development of successive soil horizons. The latter bio-chemical process is a result of the clay/humic/Fe-sesquioxides vertical transport in thesurface layers under the aegis of leaching respectively enrichment enhanced along theroots of any kind of vegetation cover. This process is called “pedogenesis” or soil de-velopment resulting in the development of a series of specific “soil” horizons in thestrict pedological sense. Depending on the nature of the parent material (hard or loose)and of the prevailing climatic conditions, different soil types may be generated whichare compiled in the international soil classification system. In reverse, from the soiltype former conditions of climate and vegetation cover may eventually be disentan-gled. Despite the fact that pedology has the state of a well developed science in earthsciences the term “soil” as stated above has still different meanings and connotationsdepending on the professional field in which it is being considered and not at leaston the skilfulness of the scientist involved. In fact numerous geo-scientists misuse thename too. And what about the scientists not acquainted with earth science at all? Thisis absolutely true in the field of astrobiology. What is then the meaning of the label“Red Soil” on Mars, Venus and other planets? In fact are they really “Soils” as defined

above or just simply red coloured (pediment) surfaces as the ones covering broad ex-tensions in the tropical regions of Brazil and Congo? In fact, soil weathering shouldbe clearly dissociated from all other types of rock alteration processes. Moreover soilcomposition, especially clay – phyllosilicates, display characteristic features enablingdetection of real pedogenetic processes. Close observation of the impact traces at theMars module site as well as the recent detection of phyllosilicates clays on Mars, maylead to firm indicators about pedogenesis processes on Mars. This opens new possi-bilities for the study of soil development similar to earthly soil processes on Mars andperhaps on other planets of the Solar system as well.

Bacteria occur in great amounts in soils on Earth. Both aerobic and anaerobic bacteriaoccur so that a great variety of species is shown. Most unexpectedly there number ishigher in desert environments rather than in moist places like the Amazone. Hence,soil cyanobacteria play an important role in the building of microbiotic crusts in ex-treme environments of drought and cold like Antarctica. Cyanobacteria also induceimportant biochemical cycles such as the nitrogen-fixation in soils. Geologically theirorigin may be far remote in time so that it may be assumed that they are time-resistantas well. Therefore soils of extreme desert conditions, cold and warm, on both Earthand Mars may contain great amounts of such resistant bacteria. Soils in the pedologi-cal sense if present on Mars and other Planets may then likely open new broad fieldsof investigation which research has hitherto been somewhat neglected.

As a consequence, now that phylosilicates have been detected on Mars the role ofwater in the soil-weathering process of clays has undoubtedly been proved. This couldfurthermore imply that not only water and soil weathering / pedogenesis extendedover the entire surface of Mars, but a vegetation cover as well. The far too generalconnotation of ‘soil’ should then be reconsidered as a true soil concept inferring bothwater and vegetation in its development.

Hence, the relationship between soil development and vegetation/bacterial life on thesurface of Mars opens up new broad possibilities for studies in astrobiology. Soils inSpace and their related cyanobacterial content should become the genuine research forall evidences of real soil development outside planet Earth in our Solar System.

Additional automated missions will most cer-tainly occur, but the ultimate scientific study ofMars will be realized only with the coming ofman—man who can conduct seismic and elec-tromagnetic sounding surveys; who can launchballoons, drive rovers, establish geologic fieldrelations, select rock samples and dissect themunder the microscope; who can track clouds andwitness other meteorological transients; whocan drill for permafrost, examine core tubes,and insert heat-flow probes; and who, with hisinimitable capacity for application of scientificinsight and methodology, can pursue the questfor indigenous life forms and perhaps discoverthe fossilized remains of an earlier biosphere.(Benton Clark, 1978)1

The New Mars

In the 1960s, most automated missions beyond low-Earth orbit—the Rangers, Surveyors, and LunarOrbiters—supported the piloted Apollo program. In the1970s, as NASA’s piloted program contracted to low-Earth orbit, its automated program expanded beyondthe Moon. Sophisticated robots flew by Mercury,Jupiter, and Saturn, and orbited and landed on Venusand Mars.

Though they were not tailored to serve as precursors tohuman expeditions in the manner of the Rangers,Surveyors, and Lunar Orbiters, the automated missionsto Mars in the 1970s shaped the second period of pilotedMars mission planning, which began in about 1981. Thefirst of these missions, Mariner 9, took advantage of thefavorable Earth-Mars transfer opportunity associatedwith the August 1971 opposition to carry enough pro-pellant to enter Mars orbit. It was launched from CapeKennedy on 30 May 1971.

In September, as Mariner 9 made its way toward Mars,Earth-based astronomers observing the planet throughtelescopes saw a bright cloud denoting the onset of adust storm. By mid-October it had become the largeston record. Wind-blown dust obscured the entire sur-face, raising fears that Mariner 9 might not be able tomap the planet from orbit as planned.2

On 14 November 1971, after a 167-day Earth-Marstransfer, Mariner 9 fired its engine for just over 15 min-

utes to slow down and become Mars’ first artificialsatellite. Dust still veiled the planet, so mission con-trollers pointed the spacecraft’s cameras at the smallMartian moons Phobos and Deimos. In Earth-basedtelescopes they were mere dots nearly lost in Mars’ redglare. In Mariner 9 images, Phobos was marked by par-allel cracks extending from a large crater. Apparentlythe impact that gouged the crater had nearly smashedthe little moon. Deimos, Mars’ more distant satellite,had a less dramatic, dustier landscape.

The giant dust storm subsided during December,theatrically unveiling a surprising world. Mars wasneither the dying red Earth espoused by PercivalLowell nor the dead red moon glimpsed by the flybyMariners.3 From its long-term orbital vantage point,Mariner 9 found Mars to be two-faced, with smoothnorthern lowlands and cratered southern highlands.The missions to the Moon confirmed that a relation-ship exists between crater density and age—themore densely cratered a region, the older it is. Hence,Mars has an ancient hemisphere and a relativelyyoung hemisphere.

Mars is a small world—half Earth’s diameter—withlarge features. The Valles Marineris canyons, forexample, span more than 4,000 kilometers alongMars’ equator. Nix Olympica, imaged by Mariner 6and Mariner 7 from afar and widely interpreted as abright crater, turned out to be a shield volcano 25 kilo-meters tall and 600 kilometers wide at its base.Renamed Olympus Mons (“Mount Olympus”), itstands at one edge of the Tharsis Plateau, a continent-sized tectonic bulge dominating half the planet. Threeother shield volcanoes on the scale of Olympus Monsform a line across Tharsis’ center.

Most exciting for those interested in Martian life weresigns of water. Mariner 9 charted channels tens of kilo-meters wide. Some contain streamlined “islands”apparently carved by enormous rushing floods. Many ofthe giant channels originate in the southern highlandsand open out onto the smooth northern plains. Thenorthern plains preserve rampart craters—also called“splosh” craters—which scientists believe were formedby asteroid impacts in permafrost. The heat of impactapparently melted subsurface ice, which flowed out-ward from the impact as a slurry of red mud, thenrefroze.4

53Humans to Mars: Fifty Years of Mission Planning, 1950–2000

Chapter 6: Viking and the Resources of Mars

Mariner 9 depleted its nitrogen attitude-control propel-lant on 27 October 1972, after returning more than7,200 images to Earth. Controllers quickly lost radiocontact as it tumbled out of control. A week later, on 6November 1972, mission planners using Mariner 9images announced five candidate Viking landing sites.5

Viking 1 left Earth on 20 August 1975 and arrived inMars orbit on 19 June 1976. Its twin, Viking 2, leftEarth on 9 September 1975 and arrived at Mars on 7August 1976. The spacecraft consisted of a nuclear-powered lander and a solar-powered orbiter. The Viking1 lander separated from its orbiter and touched downsuccessfully in eastern Chryse Planitia on 20 July1976. Viking 2 alighted near the crater Mie in UtopiaPlanitia on 3 September 1976.

The first color images from the Viking 1 lander showedcinnamon-red dirt, gray rocks, and a blue sky. The skycolor turned out to be a processing error based on pre-conceived notions of what a sky should look like. Whenthe images were corrected, Mars’ sky turned duskypink with wind-borne dust.6

The Vikings confirmed the old notion that Mars is thesolar system planet most like Earth, but only becausethe other planets are even more alien and hostile. Ahuman dropped unprotected on Mars’ red sands wouldgasp painfully in the thin carbon dioxide atmosphere,lose consciousness in seconds, and perish within twominutes. Unattenuated solar ultraviolet radiationwould blacken the corpse, for Mars has no ozone layer.The body would freeze rapidly, then mummify as thethin, parched atmosphere leeched away its moisture.

By the time the Vikings landed, almost no one believedany longer that multicellular living things could existon Mars. They held out hope, however, for hardy single-celled bacteria. On 28 July 1976, the Viking 1 landerscooped dirt from the top few centimeters of Mars’ sur-face and distributed it among three exobiology detec-tors and two spectrometers. The instruments returnedidentical equivocal readings—strong positive responsesthat tailed off, weak positive responses that could notbe duplicated in the same sample, and, most puzzling,an absence of any organic compounds the instrumentswere designed to detect.

Viking 1 and Viking 2 each scooped additional samples—even pushing aside a rock to sample underneath—and

repeated the tests several times with similar equivocalresults. Most scientists interpreted the Viking resultsas indicative of reactive soil chemistry produced byultraviolet radiation interactions with Martian dirt,not of life. The reactive chemistry probably destroys anyorganic molecules.7

Improved cameras on the Viking orbiters, meanwhile,added detail to Mariner 9’s Mars map. They imagedpolygonal patterns on the smooth northern plainsresembling those formed by permafrost in Earth’sArctic regions. Some craters—Gusev, for example—looked to be filled in by sediments and had wallsbreached by sinuous channels. Perhaps they once heldice-clad lakes.

The Viking images also revealed hundreds of river-sizebranching channels—called “valley networks”—inaddition to the large outflow channels seen in Mariner9 images. Though some were probably shaped by slow-ly melting subsurface ice, others appeared too finelybranched to be the result of anything other than sur-face runoff from rain or melting snow. Ironically, mostof the finely branched channels occurred in the south-ern hemisphere, the area that reminded people in the1960s of Earth’s dead Moon. The flyby Mariners mighthave glimpsed channels among the Moonlike cratershad their cameras had better resolution.

Low pressure and temperature make free-standingwater impossible on Mars today. The channels in theoldest part of Mars, the cratered southern highlands,seem to point to a time long ago when Mars had adense, warm atmosphere. Perhaps Mars was clementenough for a sufficiently long period of time for life toform and leave fossils.8

The Viking landers and orbiters were gratifyingly long-lived. The Viking 1 orbiter functioned until 7 August1980. Together with the Viking 2 orbiter, it returnedmore than 51,500 images, mapping 97 percent of thesurface at 300-meter resolution. Though required tooperate for only 90 days, the Viking 1 lander, the lastsurvivor of the four vehicles, returned data for morethan six years. The durable robot explorer finally brokecontact with Earth on 13 November 1982.9

Viking was a tremendous success, but it had been wide-ly billed as a mission to seek Martian life. The incon-clusive Viking exobiology results and negative inter-

54 Monographs in Aerospace History


pretation placed on them helped dampen public enthu-siasm for Mars exploration for a decade. Yet Vikingshowed Mars to be eminently worth exploring.Moreover, Viking revealed abundant resources thatmight be used to explore it.

Living off the Land

During the period that Mariner 9 and the Vikingsrevealed Mars to be a rich destination for explorers,almost no Mars expedition planning occurred insideor outside NASA. The Agency was preoccupied withdeveloping the Space Shuttle, and Mars plannersindependent of NASA—who would make many con-tributions during the 1980s—were not yet active insignificant numbers.

Papers on In-Situ Resource Utilization (ISRU) wereamong the first signs of re-awakening interest in pilot-ed Mars mission planning. ISRU is an old concept, dat-ing on Earth to prehistory. ISRU can be defined asusing the resources of a place to assist in its explo-ration—the phrase “living off the land” is essentiallysynonymous. In the context of space exploration, ISRUenables spacecraft weight minimization. If a spacecraftcan, for example, collect propellants at its destination,those propellants need not be transported at greatexpense from Earth’s surface. In the 1960s, ISRU wasstudied largely in hopes of providing life-support con-sumables. By the 1980s, the propellant productionpotential of ISRU predominated.

NASA first formally considered ISRU in 1962, when itset up the Working Group on ExtraterrestrialResources (WGER). The WGER, which met throughoutthe 1960s, focused on lunar resources, not Martian.This was because more data were available on lunarresource potential, and because lunar resource usewas, in the Apollo era, potentially more relevant toNASA’s activities.10

The UMPIRE study (1963-1964) recommended apply-ing ISRU to establish and maintain a Mars base dur-ing long conjunction-class surface stays. Doing thiswould, of course, demand more data on what resourceswere available on Mars. NASA Marshall’s UMPIREsummary report stated that “[t]his information,whether it is obtained by unmanned probes or bymanned [flyby or orbiter] reconnaissance missions, would

make such a base possible,” making the “ ‘cost effective-ness’ of Mars exploration . . . much more reasonablethan [for] the short excursions.”11

Fifteen years after UMPIRE, the Vikings at last pro-duced the in-situ data set required for serious consid-eration of Mars ISRU. The first effort to assess thepotential of Martian propellant production based onViking data spun off a 1977-78 NASA JPL study of anautomated Mars sample-return mission proposed as afollow-on to the Viking program. Louis Friedmanheaded the study, which was initially inspired byPresident Gerald Ford’s apparently casual mention ofa possible “Viking 3” mission soon after the successfulViking 1 landing.12 Robert Ash, an Old DominionUniversity professor working at JPL, and JPL staffersWilliam Dowler and Giulio Varsi published theirresults in the July-August 1978 issue of the refereedjournal Acta Astronautica.13

They examined three propellant combinations. Liquidcarbon monoxide and liquid oxygen, they found, wereeasy to produce from Martian atmospheric carbondioxide, but they rejected this combination because itproduced only 30 percent as much thrust as liquidhydrogen/liquid oxygen. Electrolysis (splitting) ofMartian water could produce hydrogen/oxygen, butthey rejected this combination because heavy, energy-hungry cooling systems were necessary to keep thehydrogen liquid, thus negating the weight-reductionadvantage of in-situ propellant manufacture.

Liquid methane/liquid oxygen constituted a good com-promise, they found, because it yields 80 percent ofhydrogen/oxygen’s thrust, yet methane remains liquidat higher temperatures, and thus is easier to store. TheMartian propellant factory would manufacturemethane using a chemical reaction discovered in 1897by French chemist Paul Sabatier. In the Sabatier re-action, carbon dioxide is combined with hydrogen in thepresence of a nickel or ruthenium catalyst to producewater and methane. The manufacture of methane andoxygen on Mars would begin with electrolysis ofMartian water. The resultant oxygen would be storedand the hydrogen reacted with carbon dioxide fromMars’ atmosphere using the Sabatier process. Themethane would be stored and the water electrolyzed tocontinue the propellant production process.


55Humans to Mars: Fifty Years of Mission Planning, 1950–2000

56 Monographs in Aerospace History


Ash, Dowler, and Varsi estimated that launching a one-kilogram sample of Martian soil direct to Earth wouldneed 3.8 metric tons of methane/oxygen, while launch-ing a piloted ascent vehicle into Mars orbit would need13.9 metric tons. These are large quantities of propel-lant, so conjunction-class trajectories with Mars sur-face stay-times of at least 400 days would be necessaryto provide enough time for propellant manufacture.

Benton Clark, with Martin Marietta (Viking’s primecontractor) in Denver, published the first papersexploring the life-support implications of the Vikingresults. His 1978 paper entitled “The Viking Results—The Case for Man on Mars” pointed out that every kilo-gram of food, water, or oxygen that had to be shippedfrom Earth meant that a kilogram of science equip-ment, shelter structure, or ascent rocket propellantcould not be sent.14 Clark estimated that supplies for a10-person, 1,000-day conjunction-class Mars expeditionwould weigh 58 metric tons, or about “one hundredtimes the mass of the crew-members themselves.” Theexpedition could, however, reduce supply weight, there-by either reducing spacecraft weight or increasingweight available for other items, by extracting waterfrom Martian dirt and splitting oxygen from Martianatmospheric carbon dioxide during its 400-day Marssurface stay.

Clark wrote that Mars offered many other ISRU possi-bilities, but that they probably could not be exploiteduntil a long-term Mars base was established. This wasbecause they required structures, processing equip-ment, or quantities of power unlikely to be available toearly expeditions. Crop growth using the “extremely

salty” Martian soil, for example, would probably have toawait availability of equipment for “pre-processing . . . toeliminate toxic components.”15

The Vikings’ robotic scoops barely scratched theMartian surface, yet they found useful materials suchas silicon, calcium, chlorine, iron, and titanium. Clarkpointed out that these could supply a Mars base withcement, glass, metals, halides, and sulfuric acid.Carbon from atmospheric carbon dioxide could serveclever Martians as a foundation for building organiccompounds, the basis of plastics, paper, and elastomers.Hydrogen peroxide made from water could serve aspowerful fuel for rockets, rovers, and powered equip-ment such as drills.

During the 1980s, the Mars ISRU concept generatedpapers by many authors, as well as initial experimen-tation.16 Robert Ash, for example, developed experimen-tal Mars ISRU hardware at Old Dominion Universitywith modest funding support from NASA Langley17 andfrom a non-government space advocacy group, ThePlanetary Society.18 That a private organization wouldfund such work was significant.

Before ISRU could make a major impact, piloted Marsmission planning had to awaken more fully from itsdecade-long post-Apollo slumber. Post-Apollo Marsplanning occurred initially outside official NASA aus-pices. This constituted a sea-change in Mars plan-ning—up to the 1970s, virtually all Mars planning wasgovernment-originated. In the 1980s, as will be seen inthe coming chapters, individuals and organizationsoutside the government took on a central, shaping role.

Mars Research Papers

Documents

Transcript of Mars Research Papers