JounuL .f Th. Btitish lntetplanetarr Sui.tl, Uol.49, pp. 443122, 19qb

THE UTILITY OF GEOTHERMAL ENERGY ON MARS

MARTYN J. FOGG Prohdbilitylleseaft* Group,c/.) 11 Ho,tdtthCau,l, Fauntain Dnt", Lidon SEl9 luv lntetnet:n|bgg@cix conpulinkco-uk.

The exploikrion of geothermxl enerrry has been absent f.oin previous considerations of providing power for selllements on Mars. The reasonfor rhis is the plevailing paradigmthal placesall ofMars'v(tcrDic activity in the rcmotepast and hence postulates a crust that is liozen to great depths. It is ,rgued in ihis paper that this view nay be truc in geneirl, but false ;n particular.Geological cvidence is reviewedthat suggeslsihat magmatismmay havebeen actjve on Mars trntil recenitimes and may heDccslill be ongoing.Thus. thc Fesence of significant,localizcd, hyperthernal arcxscannot be ruled out on tbe bas;sof rhe low meanheatflows prcdictedby global heatflow modcls.The possibility ofthe Fesence of usefulgcolhermat ficlds is further srrengrhened by observations offluvial outflows that seem to have becn associated with cerlxin magmatic cxrrusionslrnd which rhereforchinl arlavourablegroundwatcf conditions. Such a geothcmal energysource would be of geat poienrial economic value, bcing of use for the genef.ttun ol electricity and direct heating lbr industry and habilalion. The addition of this energyoplion to rhoseofsolar wind and nuclear,cannot but enhancethe prospectsof a marlian civiljzatbn that must srartafrcsh. {,ithout an equivalcDLLo the Earlh's stocl of foisil fuels.

Au thinss aft an ertha .seJorfrre, and lirc lor al.l thins', evn as wdr?slbr sold and sotd Jb ware'. Heraclifirs of Ephesus

I. INTRODUCTION

Mars pronises to be both a genercusandchalie.sing placelbr sible by thc net energy providcd by the combustion of our futurc exploers to visil. Generous,becansc il is the onlyhown phncl's rich stock offossil fuels. Oil, coal, naturalgas, and rhe ext.atenestrialhcality possessingan acccssibleabundance of oxygen-rich atmosphcrc in which to bum thcm, are not present just the matcrirl pre-requisilesof iife; challenging,because ihis on Mars- ln settling Mars thereforehumaDity will not be endowmcnl |empts us ioward a lNgerlenn strategy where conlionied by the alicn rigours of Mars ilselI, bui will have to nranned visits evolve in() a pemnnent and self sufiicicnl faceihe challcngeof providingan allemativeenergetic basis for civilisatnrn. The validity of expk)ittug indigenous relourccs on Mars to In thispaper,proposedmcthods of generatirrgpowcroD Mars supportmanned expk)raLory missions is now widely accepied. nre bicfly discussed heforc embarking on a morc delailed Using matian sunlighl:rnd wind to gendale auxiliary powef. discussionof the potentialofnartian geothcmal energy.Con mafirn volatilcs as feedstocksfor lifc supporisystems and the sideraiionofthis optbn hasprobably beenncglecied untilnow maDufactur oftuels, and nrnrtiansoils androcks lbr construc becauseof the prcvailing paradignroi Mtus rs a geologic{lly ri(n, will enhancethe duration, flcxibility and safety of n;s deadand deep f.ozen world. Howevcr.it will be arguedthat this paft'culxr. stunsto Mlrrs t1-31.This. howcvcr is'living o1Tthe land" in a viewpoint. whilst true in general.may be false in restrictedsense. since it is only madepossiblewith an expcnsive Hypefhennal arexsprobably still exist on Mars and night be teclrnological infrastructure manufactured on and supplied profitabl] "rnined" for their hca1. liom the Eafth. For nrfftian settlements to prcgress lowards self sutriciency as they musl do to bc rcalistic at all this 2. OPTIONS FOR POWER SUPPLY ON MARS depeDdcnccon a net subsidyfrom thc Earth nusteventually be eliminatcd rnd replacedby a rcl,tlionshipof two independent Terresrrialcjvilisxlion relies on four power sourceslor almosl tmding econonies t4l. aI its consumpttunof pdmary energy.These are. in descelding -eqr A fJrd.urentdt rrcnenr,'fJr ) eco )5rernor ecun.im\i. order thecom buslion of fossil fuelsi tbccombustion of biom,tss an internal llow ofenergy.In addition,for such systens 1()be (mostly fuelwood); hydroelechic power: and nuclear fission. viable over the long tem and fbr growtl to be possible, energy SomedcLails of thesesourccs (applicable to lhe lxlc i980's) are resouces must be hanressedprofltably that is, nore usetul giveDin Table l Also listed area variety ofrltemative pdmarf alJ) energymust be liberatcdttl:n workconsumcdin theFocess of erirf) \uurce\!hrl 1Ir.r"l\ errber r ninrr or oon liberatnm.Thns,tbcre must be acontnluing suPplyof,el.rr€r8) cxislent, e,ronomicrolc, but which are uidcr consideratlonas andthe higherthc net energyratio (usefulcnergy provided/work futurc replacements for fossit tuel.A rough impression ofthe net e\prnJrd in pro';sioo'rfe herlcr.5'. Nerenerg) pro\i.i. n i. energy provided by thesesystems (wbilst ignoring argunrents thus crLrcill to ihe lbasibil ity of thc setilement of Mars ^nd is fxr over hidden subsidiesetc) can bc gxined by looking at the from being arrivial questioD.Mantind only hasthc cxperience quoted cost of the generation oI electricity in general, the of bujlding one technohgicrl civilisation our own rnade pos- cheaper thc price, rhe grericr the relevant net energy ratio. TLBLE l. A Conparison o.fEnery! Sources(Iate t9S0,s)

Energy Source TJpicalCost of rypical Unit Sizes World Primary Applicationto Mars: Electricity Ceneration Encrgy Use

500 1000MWc per large 15.6Io No

Hydfopower -39lkwhr 1000MWe perlargc dam s.5% No

Bionass "5dlkwhrr < 100MWe 14.11o' No

Nnclear fissbn 500 1000MWe per large 4-l9a

Tidal energy 0.4 240 MWe pd hJrragel Negligible No

Ocean thennal energy 12-25clkWtu l0 - 100MWe perlargc planC Nune convenion (OTEC)

-9dlkwhr Solar ihennal 10 100 MWe l]ef large planC Negligible

25 35d/kWhf 0.03 6.5 MWe pcr large Negligible

6-15q/kWhr 0.1 - 0.5MWe pertulbine6 Smrll

Geotherm.tl .l 109/kwhf 10 - 100 MWe per large plant O.19o

5000MWe per l.,r!e srrellire Yes sarellite(SPS)l

4.',,, /J B ^it d iJa t, .8.4,td:q.\., opo4taL, ! P]z\ert pilai pldnt! aE Dortlr af ldt.r upa.:ir^. (nat yt ' trittztt ptn ti.el). Erptttut to fdll to lt-t 5 dkwht bt 1A(tututt ].\|s iato the 2tsj Gntury 1 M.elt ih.]iL ietu r J,elvoal .onhunnL in lLnlw.n.t.:oun iet wihd lanlL\ onti,, af d tunho 4 :ach turbitu\.

Dda n st,\.rllallnn Ref t6l.

When consideringpower generaiionon Ma|s. we lrc frced estimate lbr lhe per capjta power consunprion olapermanently wilh a doubleproblem. Firstly, Mars does nor havea nalurat, eslablishedmartian popularion mighr be ardved aiby assnming uncontained,bnrsphere which provides sfdrts life-snpport. : dc n,,ndfofeleLnir:r\ ,iTilar o rhepur. Jpiup.;&n rnerg) Altificial lite-supporr, wherher entirely rncchanicat,or conslrmption of the fiist world narions, multiplied by a factof , b.oregcrerrtire.i. ,, omple\rad eqeG)inrer.i\e t-oLe.. rellectiDg the need to synlhesise alrernarivc energy carriers. and Estimatesof rhe electrical power consumprionon Mars to incrementcdfurher due lo rhe needsof lrrificial life-support. supporthu'nan beings vary from on p.ipercaiculat;ons for a Such a calculationwould enrbodyihe Deedsof basic survivxt_ tcmporary 8 manoutpost which obtain a figufeas low as- i.2 jndusty and a faif siandardof living. Thepublishedfigure rhar kwe/person i7l, to theerpericDce ofrunning Biosphere 2 which perhaps comes closest to realily is ihe calculatedelectdcal needs - anenornous l00kwe/pe$on[8]-This compares tothe power demand of the 150-man Mars base designedby the - il kwperson - and the 3 kwe/personconsumption of JapancseOhbayashi coryorarion, inrended specificalll as a primfiy andelectrical energyrespcctivety of the avcraseNncn beachheadforpemane.t seulemenrli0l.Although thcircalcu- can.The rcality ol livingand prospering on Mars will probably larions werc not tull) displayedin thc cited reference,their involvea per capitapower constrnption berwecn rhese t-100 esiimrle of an all inclusive power denand of - 20-50 kwe/ kwe/person extremes.The former is muchtoo low 6 rrqoes nor personhas an airofplarsibility- ffwe assumcthis to be realislic. include rexlistichearing re.tuiremenrs and the denrandsof thenwe mighl .easonablyexpect a g.owing civi lisationon Mars pernanent habitationand growth. including rhose of indusry needrng a supply of electdcal encrgy abour 2-5 rimes ds gear as and qxalily the of 1ife.On the otherhand, we might reasonably theconsnmptionofprimary energyof the firsr wortd narionsof expecttuture enclosed habitats to be mxch moreeconomical the Earth. and practical than Biospheft 2. which was nor designedwith The secood problem inherent in the energerics of marian efiicicnt energyuse in mind. seillcment!is that, iookirg back ai Tablc I, it is seenthar oniy On the Eafh, direct fuel useis economicallyvery imporrmr. one of the big foul power sowces is likely to be avaitabte. On Mars,it is not an optionand fucls andoxidiscrs musr be namely nuclear fission. To be broughr inlo operation, nuclexr synthesisedfrom basic raw materiais13.91. Thus, a rcasonabte power requires an elaborate inffastrucrufe in place. Nrclear rh. Utititr 4 c?.thlmat EncB ar Mat.l fncls mustbe prospected.mnrcd, punfied andthen consumed rn (1) A powerfrl sonrceof natual subte'raneanheai reactors that are themselvcs the products of sophisticatcd manutacturing.Spent fuel mxst be properly disPoscdof, or l:2) An adcqrale water suPPlY reprocessed,furthef.tdding ro the rvork involved.Thus,in order to commence thc prcgressiontowards self suflicienl energy (3) An aquitbr of permeablercservoir rock use, martianscttlc|s musi have accessto morc inelaborateand avril.tblc cncrgysources, before large scaleindigenous nucleaf power cones on line. The lwo power sources often mentioned in this fegard afe On Mars fiough, itcm 1 is particulddy problematicin view solaraod wind power L2.ll,l2l. Howevea whilstthese sourccs of the prevailing puadigm of the phnet bcing cooled to great are lndoubtedly availablc on Mars and might se.ve wcll 1() depths. Theoretical models of tle mean global geothermal lreat Fovide auxiliary power for outpostsorpionee ng sc(lements, flow on Mars 116 181eslimate it to be aboul0.03-0.04w/m', lhey afe not suitablc .ts the eneqetic basis ol .r vlgorors compared ro an average of 0.06 0.011w/nf on the Earth- civilisation especirlly in view of the 'nuch grcalcr relatlve Looking at thc problemfrom thepoinlofview ofthis meanfbw demandforelcctricity. The pricesgiven for solffrhennal. solar- does indccd render nartian gcothermal power an unlikely phobvolLaicxnd wind power in'1j$lc I show thatthese systens prospcct,as ilplacesthe 100"Cisoihem asdeep down as 10km. (espcciallysolar) are rolfully conpclitive withtbssiltuels due cvcn al theequato. Howcver,heat leaking out ofthc Earthis far to lbc high cost of marufacturing geDcr.rtingequipmenr. such ir-s from isotopic and exhibils enoflnous variationsbetween 10- solar cells and wind turbincs. Priceshowever are falling and calities.A modelthrl predictedmercly ihc meanheat llow ofthe should contirue to do so in the ne)tt century. but even if solar and Eafh to be, sry,0.065 W/mz would Dolpredicl the existenceof wind power should becomeeconomic on the Elnh i1 doesnor loca1"tows" in Ibrffrc rcgions ofjrst 0.03 w/m2. "highs" of mean they will beconc so on Mars, wherc additbnal factors ffe 0.09 W/mr in backirc regio.s, {nd "s perhighs' in imnedixtc againstthem.Martian sunlightis ar averxgcof,t39, as iniense volcanicxrcas as high as0.6w/m]1191. Even thoughMds does as tenestrial sunlighl, and even when lcsscr cloud coler is not possessptate tectonics,ils hislory has been dominatedby {ccouDlcdfor sohr arrayson Mars wili needrougHy doubleihe volcanism.Rocks of igDcousorig are thoughl to be exposed xrea as a snnilarly rated one on E.llth. Since lhe ma.tian ovcr some5870 ofits surflce not all ofthem datingback to the alnosphereat the planet\ sudaceis only about 17' the density prinordial times 1201.11is perhaps drereforc no1unreasonable as lhat or Eallh. a sinrilafly rated wind turbine on Mars musl to propose lhat nxrlian heat flow t(n might exhibit some local haverotorbladesten timcs thediameier astufbines buill lo wo* variation and that, given the possible abundance of subtenx on the Earth. or must opcrate in winds dlat average 4.6 lines the neanwater [2] 24], exploitablegeolhemallields might stillbc speed.ln rdditbn, bolh sotarand wjnd pwer arecxlegorised as beirg "intennittenr renewables" thcy never run out, but are Is it rcasonablethereforc |o disnriss the t'otential ol geothermal only intcnnittently availableand arc Lhusunsuitedto provide a power production o. Mnrs out ofhard? Noi Dccessarily.This base-loadpower supply. paperthus proceeds1() assess the possiblecxisEnce ofpresent wllat ot' the oiher power supply oplions listed as alpfofd.ttc day hypedhernal .treasonMa$ and the luture usesof matian fbr Mars inTable l? Deutcnuntis abnrdanton Mal'sand so there georhermalencrgy, should it tum orL to be viable. exists plenty of fuel Ibr nnclear fusion. Bur lusion is an unprovencorcept and wil likely face the samebootstrapping 3. POSSIBL!] HYPtrRTHERNL{L LOCAI-ITIES problemssuggestcd previously for fissbn. Solar Power Satel- ON MARS lilcs (SPS) will be suitable fof thc provision of base load clcctricity as they canbe sitedin pcmanent sunlightLl3l. Bven If ir were easyto drill boreholesto an {rbitnry depih,lhen hot given lhe weaker sunlight at Mars, the OhbayashidesigDc.s rock could be reachcd lion anywhere on the surtace of Mars consideredSPS as the bestmelhod ofFoviding iheirsettlemenl However economicand Factical confainls dictatethat a good with a rcliab1epowe. supply J101.However. SPS is also an geothennalficld is underlainby relxtively shallow.hot strata unFoven concept(although cerlain to be viable) and has rhe This requnes an anomalouslyh;gh heal flo\ implying thc disadvantage of bcing sited remote frcnr thc se[lenent, in p.esenceofa young igneousinlrrsion beneaththe site a body of ascending magma thai nray or may not ultimateiy rcach the This leaves iust one remainjng contendcr il1 Table l- sufface.A crucialquestion to asktherelde k whethermag atic geolhermai energy-heat enanatirg frcm the planetary interiol ictivilyis still extant:for themore commonplacci1is, thebetter and fbund in usetul form as hot nndeigronndwaterc. This has the prospectsfor geotbcrmalpower productiot. five attactive features: it is an inlrinsic eoergy resource: it is Theoretically.anywhere on Mars could be underlainby a snilabletbr a baseload lupply; in additionlo electricitygcnct? liesh magma body but, as on dre Earth, most of these heat tion. it can be used to pro\,ide thermal energy dircctly; the sourceswould Do1naniftsl dramaticallyat the suface. The lechnology is compa.atively maturer and. if a quality presenceofsuch'hidder" sourcesis often diflicult to infef in hytuothermalrcsefvoir canbe found. then power is potentially the absenceofdetailedon-site obscNalions. However. it 'nay bc very cheap to generatc. Experience has shown that gcolhei al ihat our currenl data sel can act rs x rneaningtul guide to whcre electricpowerphDls ltre exrremelyreliable md nexible, being at least some present day m.tnian hyperthemal arcas nrighl be on line 9770of the time, comparcdto 757, and 657. for coal silualed.This is becatrsethe most obvious ertenal manifesta- fircd and nuclearplanrs respectivcly. They can be conshicted lions of internal heal a.e volcadsm and possibly oulbursi in modular increments and hencc upgraded in capacity when flooding-and featu.es rclating to both of thesc processes have necessary:0.5- 10 MWe planLscm be np andnlnning in asshon been eritensjvely nnaged from orbit. It follows theretde that ll as six mondrs and clustered iaciliries generating > 250 MWc in searh.for hyperthemal arcas on Mars shoukl st.ltt br emmin just 2 yeals t l,ll. i g theplanetkyounsest igneout andflwial tentin, assunins However four conditions must pellain ir ordcr lor there lo rhat some o:f their nas atic centtes md! still be actiw or exist an exploitablcgeolhennal field (ignod.g, for themoment, .lornant, as oppose.l b long extinct. js the pot€ntial ol hot dry rock): theseare [] 5l: Martian geology currently divided into th.ec major chronostatigraphic divisions, the Noachian. Hespenm and ties. but before revicwnrg these it is lsetu1 to consider the Ana:ofri?z Systenrs (in order of decreasing age). These are planct wide distribxtion of such areas. This can be done by ftflher subdividednrto rowe.ald U/rp.f. and in ihe caseoIlhc refhring t{) the USGS 1:I 5,000.000-scalegeologic lnaps of Noachian and Anrzoninnl, Middle Epachs. Il/4t{ Ana.ouan Mars prcparedby Scott ar 1301.These display a wide variely (UA) "r surfacesare datcd as the mosrrccent. having a superposed of geologic units which are conelated by iheir cratef frequency densityof impactcralers > 2 km-dianercr of lessthan :10 per 106 Jinf.burion.ru he hreen n( s).,rrn\ in an :rJ- - "Idrigraplxc km:. They cover 7E ol thc planer'ssurlxce, witt 32E ;f them companyingcharL.Akey describing map units (which arecoded characterisedas being of volcanic and 127, of tluvial orjgin borh by coloul and I fonnal lener sequence)is also pmvjded. - I20l. This implies that 3.i7, ofthe sudaceof Mars (anffcx of whicb gives a generalinterprctation of their composilion and - 4.5 million klnr) is coveredin "young"igneousrocksorftuvial odgin. These naps therefore give an oveNiew of the areas of sedrments.This is a subslanrialarcx, comparable in siTetohalf inleresl herc and the relevmt data is sunrnarised in Tabte 2. Ir the USA. is apparentthat UA volcanic and fluviat rocks (, ,,o! occur aL Daling a sudaceon Mars is doneby counringthe locntcrater random. ln fact they are almost exclusively k,cated in rhe densityand then calculxtingthe rine rcquiredto seDcruemose planetk northwcst qua&anr, hom lorgitude 220' in Elysiurn craters by using an irnpact flux model scaledretarivc to the eastwafdto longilude20'inAcidalia Planitiaand no.th of t5.,S Moon. Conpxirg craterdensities can give a good imFession and south of 50'N. Such is the clusbnng of such ourcrcpsin of the relaiive age of various sulfaces.bnr unfortunarel],Lhe adJaccntgeog.aphic areas thar one can sunnisethe ddsteDccof model dependenlpa$ of the procedure leads to substantjal a distinct province of reccnt anomaloushext flow on Ma.s. vanatunsbetween differenttimescalcs in estimalesof absoture including Ellsium. Amazonis Planitia. Arcadia Planitia and age.Thus, althoughUppcr Anazonian suffacesare thc young- Thnrsis. est (in lact Mars can slill be said ro bc within the UA). rhe absolnie age of its juncrion with the Middle Amazo.ian is 3.1 Cerberus Plains rnceltain. The Neukum-Wiscchronostratigraphic nodel dates thc UpperAmdzonianEpoch bctween 0 - 700 million yearsotd One of the youngestcgions on MaIs showing evidencc of l25l; whereastlrc Hafnann-Tand 106 Resolvingthe ageof r given marrirn fearue ro wilhin the time kn: fiom longitude 220" easiwmd ro 1680(an cxsr wesr dls- span of this Epoch is eveDmorc problernatic.especially ifthe lrDce of almosi 3000 km) and are up io 700 km widc (no(h- feature is lqr sma io provide statisncally rcliable cratering souih) near longitudc 195'. They are alnrostuncrarered and in data. On a ftrmation thal is already very sla.scly craicreol low to medium resoluLn)nimages appcf snrooth,wirh bbate uDcrateredpalches of essentially zero age xrc difficult, o. albedo pattems and embayment relations wirh older telTain. irDpossibleto dislinguish. Whaiever ir wrs that re-surtacedthis a1€a.ir appoarsto have This meanslhat cvcn if UA rocks afe "young" nr a relative filled a topogrxphic low. cented on 5oN, 190.W south of rhe sense,rt is possiblethat Dosrofrhen could stillbehundreds of Elysium volcanic assemblage and rhen o!clftowed nxlions of yeafs old. On the other hand, northeastwardsdown channelsncxf lloN. 179oW onro cont.npofdneous originlbr sane of them cannat be nle.l aut. Indeed.it seemsunlikely thal .ll nragmaricactiviry on Mars ln thc USGS naps, most of the Cerbcrusplains area is would haveceascd injusrthelasr:l l5% ofihe ptrnet.shisrory_ designatedas Unit A.re "younger channel and flood-piain This point of vicw receivessonlc suppor ftoin the evidence naterial." However. a more dcrailedstudy by Ptescia(which providcd by the SNC ncteofires (assumingthey do odginate, also includespffts of adjacentUnits.lpr andApt) suggestsan viaimpact ejecrioq liom M{rs). Thoseofthe Shergouilcgroup odgin by flood volcanism 131.321.High resolutionjmages of appear io have crystallised from la\.as as rccent as 160 Myr to ceftain areasshow distirct lava flow textues such as pressurc severalhundred Myr ago f27,281.ln otherwords, rhey are Upper ddgcs. festoodng and digitate flow margins.Plescia explanrs A'nazodan rocks quire possibly much younger than the UA lhe associatedwater-cut lcatu.es io the easl by suggestingthal MA boundary-An ageof i60 Myr is equivalenltojusr 3 5% of the lalrs floodedinto westemAmazonisvia channelscarved by manian geologic Lime.The possibility that ma ian rocks this a previous fluvial episode.This fluvial episodeand the subse- young 'nighl have reachedthe E rth by such an jmprobable .tuent volcanisD may have been crusally linked_ roxte .t.guesstrongly, on statisticalgrNnds, both lbr the likeli- The thichcss of the re-surhcnrg material is substantial. hood ihat magmatic acliviry is ongoing aDdrhar irs p.oducts Judgingfromthedisl burionof older rc.rainprotruding thrcugh mny not bc exceptionall] .are. ln a reconsrrucrionof the thc deposits,Plescia csrimaied rhar fte ccntralflows musLbe ar resurfacingrccord of Mlrls, Ta.aka s/ dl. f29l concluderhat the leasL400m ihick, fiinning ou o - i0 m Lhickindividxal llows Lcofresrdacing fiaslallen ovcf rhe iotal spanof the ptancL.s towalds the margins.This anDuntsto a visiblc releaseof - l0r - - hislory fron I kDlyfto 0.01 kmlyr. as volcrnism changed krnr oflava and - 10'6kg olwder, assuminga l,/,wr contenroI tlom widespread 10 krcal acriviry. Howevea given the avenge water within the extuxdednagrna. It is likely thaLmuch more UA rcsnfacing ratc, rnd rhe facr that lava llows oD Mars are nrelrthan rhis wrs generatedand nrtruded at shaltowdeprhs.The usxally biggerthanonthe Earth,rhc] estimatethara substanrial wateritself addsup to a - l0 m deplhavemged over theCefberus tava flow has been crlruded somewhcreon the planet once afea perhaps sufiicient ro explaiD the associalcd fluvial ever) . In ve.r. lor tr( p.r,r eq buodrrJM)r 'l'he gco'noryhology to thc cast. Plescia suggeststhat the lavas implications of the Upper Anazonian geologic rccord ernplcdat fiighratesfb.lairly briefperiods,fed by risingmnrle arcclear:MaN,r a lllore qricscenrplanetdmD the Earih,buL nor plLincs. ScalingAonl ratesof flood brsail crhrsion esrimalcd lotrlly so. Local magmaticactivity. extrusiveor mercly inrru lion thc Earth and Moon, hc calculared thal Lhc total iime fbr srvc, nlay still persist. As Tn lka d dl [20] have staled, emplacemc.t of the entire fbnnation coxld havc beenas shof "Ahhough suchpmccst6 hor" ||dned durins ttE rast 2 GrL. as 7 years or as long as,l nillnm yeals. This assunredone Juultins, wlcanisn an l Jlootlins mar dll occ0 h thefutun." continuouseruption. ln reality thc.c was Likely 1(rhrve beeD With aviewtoidcntilringrecenl gcologicacti\,iry,a number interniptionsitr activity and periods of quiescencerhat would ol delailedstudies have bccDdone of Uppef Amazonianlocdli- have further drawn out the process. The Lhi\b.lG.oth. nl En.rar on Mars

TARLE 2: Upper Anazonian |ok:anic and Flurial Tetfttin

Geologic Asscnblage CraterdeNities>2kd

Ccnndl & $uLhem Arcadia Planitia: -24 Dark, liesl qpeanng tava centrdlAcidrlia Pladtia llowsi smdlllolcanic cores.

Ccntal & southcm .\cad,d Pldnni.r 20 40 Ddk, ftesh lplcadug lava flows: smll volcanic cones

SotrLhol OlytnpusMons. wesrFard 20 50 Pyn)clasli.deposils.or.eolian alongsouthem ADrazonis Planida. -20 South& soulhea(Elysium

Sonth Elysium -2n ntfui lhanncl!\ren \L No.th edgeof lemle Fossae: northern Arcadia Planlli!.

Pl.lns flanrins Arsia,Pxvonis & 20-4n Lala llows liom lissures. AscraensMons and caldera l1l.

Pl.in flarting soudrcaslof Qllmpus 20 30 Lala flows fron fissuresi isolated fluvial features.r

Olympus Mons sheild. l0 40 'Wester! fla* of Hecates Tholus. < 30'

' lhelts(itnaps!1.n.td tinsu^h t.tnh||nh a>2km.ntrdeiri4 ol<20/ltt kn|. Sanc rc|on: inthisTuble.tuninuall!un rdture.l )nEc!hn5f1uifsdt.t.1disti4ui'h.na:an^r!?nbla!.i|||heUsGsm4Ps'hi'fh.rh&u*.Jid]tdMkJ1nd!u|L"|Da.l.lP1efi.[3]] j Tht!. rh.h.b dE t,a c.,rya2d $ith thr:( .1 lh. c.jncrus I'lni4s. "No|inc|ud.dintheU'osna?:'Th.Le.al.s7holr6danei'oJHesPerinuge|U"i|h.,'b|tMau3nri|Mdelul|33l.1ajna4UA tD{laftnlctut.rh3iatisticsBr,ennl[3J],hr1u!(llihtdlctldstDuryerthanOttnPL!MdttaLd.td.

The souces of lavas in r flood basall provlnce are otten 3.2 Ilecates Tholus dilficult to find a-sfissufc venis are buried by theif own lxva. However Plescia concluded that the geomorphologyof the HecxtesTholus is the nothernmost oflhe three large volcanic westem pat of Ccrberusnear 200'W changcd1o being more construclsof Elysium and is thoughl lo be the oidest.Its shield akin to "phnN slyle" volcanism.ch{mctcdsed by a sequenceof is nrxppedas Unit Hr?t (Upper Hesperian)and i! thus roughly ovcrhpping tube-fedlow shields.cxtcnding over broad areas. 2 3 billion years old. Its southernflanks were buricd about a Hcri:, he lvas able to nominatecc.tah featuresas being lava billion years later by LowerAmazonian lara flows fron Elysium sourceconstructs. but only on inages widr poof detail due to Mons. At face value therefore. we might reasonably expect low resolutionand high sun conditions(see Plate l). HecatesTholus to be long extinct and cooled io considerable The Cefberusphins are nndoubtedlyvery you.g and they overlieUnitAdr of thcArcadialbrmation in the casl(irselfvery MotrginisMark pl rl t33l howeverhaveihawn attentionto sparselycratercd). Despile rhis geologicalyouth howevet ii is ^ remark$le asymmetry in thc distribution of craters on this only possiblccunenily to datethem in absoluleteins to being volcano.Tothe wesi of the strmmitanddownslope sonrc 75 kn, yotrnger ihan 250-700 Myr. If they .rrc only tens of millions of there appearsan almost compleie absenceof supcrposecllmpact ).rr\od.rl'enrr r.po,.rhlerl-,r l-r.,1 llos n | |.areJtem rn. cratersand an obscurationof lhe sinuouschannels that cover the substantialy elevatedover tbc global mean.ln fact, it may be rest of irs flanks- This re-surfacing is cvidcnlly very young, that the activity undcf the Cerberusplains is not cxtiDcl, but younger it is suggestedthan 300 miliion years-lhe estimated merely in repose.The only fedures lhat areseen to posLdale the date ofthe collapseof fie Olymlus Mons caldera(Unit Arr). lava flows are the CerbcrusRupes-a wNW rrending fraclure The explxnaiion may be that wc are observing an nir fall system.Jn this rcga , ir is worth noting that grabenlbnnaiion deposit of volcadc ash, enplaccd fblowing a I'linian style due to crustalexlensiononthe Earth(which may ormaynoibe explosiveeruprion. To haveobscured lhe pfe existing topogra- alrlogout is olten associatedwith magrnalicaciiviry phy, ihe depth ofthis pymclasdc mantle nNst bc aLleast - 100 Il the Ce*erus plains afe of volcanicorigin, then it demon n andimplies avoluneof.rshof- 65 kmr andan ssof- 7x10rr strateslhat late irmartian history suflicieni heal remains in the kg of a similar magnitudeto the massesofair fall depositson mande to generatesurface hotspols ofconsiderable extent and theEar1h. ln a dctiiled modelofthe proposederuption, Moug inis nassive outpournrgsof low viscosity lava. The fact tbrl the Mark .t aa prcposed that fie a-shcould have been emplaced in outpouringof theselavas was also associated with thc releaseof a single 20 30 day eruption. of .r number of shoder evcnts watef rdds to thepossibilityofuseful geothemal ficlds remain- sepamtcdin lime by < 0. I Myr If walerwas dle driving volatile, ing xt the sire to this day. theDlhe mininum depth oi magma storagewould hrve been Pr.t ol drc {eslern exFnsc of tlD Cerberus Plains,slnNtu8 albedotxrLemssregesrive oi recenl |lva iloi! oler wide &as, subnrergiig a more ru!:ged tctrah benedlh. Thc brighr circul.r fearrre with mdl.dng llow\ to ttie wcst of the inselbeg in lhc left ol Lheliamc (ma.hcdbI rn aroiv), hasbccD lnterprered b] Plescia llll cs . lo* shicld vent. viking frame: lE5S.12rat 6'N, 207.W: imagc width: 175 kmr resofuLbn:231..1lYpixel. Phato .aunc\t 0J NSSDC

PtaitN Uplands bonndarl in thc region ot l"S. \,i 157'W showingsmooth, amosl unc.atcrcd,Uppe. AhrToniin tefunr bwards the rop ol ttrc iDa8e. nrleQretedbI ScoLrmd Tanak [3:1]as ignimLrritc ).,. delosits. They suSgeslLhe depEsslor m.rked hy thc arrow nray represenl d collatsed tnagma chambd benealh a source renl. Viknrg frLe: 606A62r ldage widrh: 750 km. Phata conei. of NSSDC t h. Lnilir^ tf G?.thlnal Enett:l at Matt

200-.1000m: heat conductinginlo the adjacentvolume could release.Asactivityh{s coillinxedfor so long. {nd.ts someofthe have meltedany grcund icc present,selting up a hydrothcmal mosr subsiantialigninbriGs appearto bethe youngesi,it is lar circul{tion th{t could have incorporatedexha walcr into lhe ftom cenainth.ir.ill magmatismbene{th thisregionhas ceased. Thecentraland easternstreiches of thc Medusae Fossae fom{ The cxlrrordinary implicationsof this obscrvalion,and nore tion might thusbe a good placeto prospectfbr anomalousheat subtlc !llrirlions in craterdensity over thc wlrole shield.is that flow, sra.dig in paficular u,ith those caldrons that hint a1 HccaresTholns has been intemiilcnlly active lbr in excessof magn chambers beneath. hundredsof milions of years. Altlrough this is unknown for volcanoeson Eardr. it is perhrps possibleon Mars where the 3.4 NorthwcsternTharsis absenceof plate tectonicsfixes lhe crust in place over m.tntlc holspotsfo. as loDgas thcy last.lllrlay be t|erefd e thatHccalcs Nowhere else on Mars is evidence fof pcrsistenl volcanism Tholtrs is not exlinct al r1l, but dorna.t still, aDdhot within. more spectacular ffrd widespread than in the Thalsis regon wherc igneousrccks covering> 7x106kmr have beenaccumu 3.3 Medusae FossaeFormation lr InSra"re '{ lr"' cor'nually .rn.r'l ppcr He\pefranrtmP.. Exlensivc lxva flo$'s flanking thc shieldsof the three massive Apyroclastic orignr forAmazonian dcpositsof a much greatel volcanocsof the Tharsis brlge, A$ia. Pavonjs rnd Ascraeus cxLcnlhas been proposedby ScoLland Tan*a 1341.The rocts Mons aredated as UpperAnrazonianin age(Unit,4t ). .tsis lhe in questionare those of ihe MedusacFossae lbfl ation.expoied malerial filling their crldcras 1301. along ihe bofder betwccn the highlands to the $uth xnd The vefy youngcsl fealures in ihis rcgion however are Amazoris Planitia witlrin an area roughly delincarcd by a thoughtto occur in northwestTharsis.associated wiih Olympus triangledrawn berween ihe volcanoesApolljnaris Palcra,Biblis Mons. The shieldofthis nassive conslruciis alsoof the Upper Pateraand Olympus Mons. They conlist of smoolh or gendy Amazonian (UDirAoi), but more reccnlly fbrmed than this arc undulatnrglevcl sheets,thai in some areasof appa]entlypoor considercdto be the plains at thc baseof the Olymplts Mons lithificaLion,show signsofhaving beencroded by thewind (see es.arpnrcnt(UnitAop), surouDdingit Aom the noftheast1() ihe Platc 2). FeaNresdiagnostic oflava flows are absentand Scott sou&. These have been inlerFeted as consistiDgof nany :nd T.uaka proposed ihat these focks are ignimbrite units ovcrlapping lava flows, rggregating to a depth oI several sheetsof padially welded volcanic ash and punice resulting kilometres. extmded frou faults and fissurcs beneath the vol fiom recuffent explosivc crupdons- cano and appear to represent the most rccenl hrge-scale fbrma Theseeruptions nnst havebe€n on an enomous scale,even tion in the regnrn. whilst it would be wise to eventualll takin: into accouDtthc lrcl thal pyroclasticcbuds shouldtravel invesiigateTharsis as a whole fof its gcothennalpotential. thc frrthef on Mus than on lhe Earth before dcposiling their load. pl{ins comprisingsequence Aop .tpp*r padicdarly promising The thicknessof the proposedignimbnres which folo\l but on two counts.Not only do they provide a clue to wherehear subdueunderlying ropography. is estimaledtobe as much as 2 flow was highest in tbc recent past. but thcy also show 3 km. Their total area covcrs - 2.2n10! kmr, comprising aD geonorphological cvidence lbr the releasc of what may have estimaiedvolume of - 3.85x10i'kmr of pyroclastslHowcver been hydrot|ermrl fl uids. MedusaeFossaerocks vary quitewidely jn ageand arcdaled to lsolnted,fluid crt, channelsin lorthwest Tharsishave becn throughoutthe Amazonian System.Only abouL.t thnd of the studied{nd describedbyMougnris Mffk [35] who considcrsit exposures.rppcf Lobe upperAmazonian in ageand aretbund nrosl likely thal their origin was due io the erosive actiur of predonrinNtly from lhe centreto ihe eastof the assemblage. flowing waler.rather than lav.t. Thte such small fluviaL oul- In their detailed study, Scoft ,tnd Tanaka subdivided the flows are located on Unil Arp and thus po(drle even this igninbrites into seven Unils b.tsed on a number of youngestofTharsis tenxin. The best imagedsyslem is located geomorphologicalcriteria. Units 6.tnd 7 (ofroughly equ{l arex) at 16'N, 129'W withnr 15 km of the OlynrpusMons escarp being the UpperAndTonian ciposues and hencecorrespond menr-and appe{.s to originale from an arcuate graben (seePlate ing to ihe USGSm.rpUnitAn,. The rocks ofUnit6 arc scento 3). The channclsystem itselfis 85 km in lengthand divides into overlieseveral of the aurcoledeposits of OlympusMons bu1rre threemain segnentsof width 300 I600 m, with bmided floors embayedby youngcr plains lava flow!. Thosc ol Unit 7 how- ard streamhredislands. Some distribuiariesend in deltasand ever .Lppearto bc tu young-or younger than the mosl receni what may be sint holes or ponds where water presumrbly Tharsis lava flows, as wel as being sone of the thickest ev.ipomtcdor percolatedundcrground.The channels secn 1()be igninbriles in the enlire formxtion.The two largestoutcrops of .tuitc shallow (- 23 m mcan depih) and the sudrcc ffea of lhe Unil 7 (tacts of land over 300 kr across) afe located at the enlire sysiemis - 50 km'],involving dre rcmovalof- 1.15km' intersedion due south of Olynpxs Mons and west of Biblis of sudacematerial. The estimatedvolume of wxler requiredto Parera.and fue eastofNicholson cfater.li is interc(;Dg to note causethis erosionis 3-12 k]nr. that both of these arcas sho$r the presence of ebngale collapse Such gcologicaly recentfioodnrg is somewhatpuzzling as structures deprcssions that Scott and Tannka suggested may models th{t predict the presence oI water on Mars today from have fomed by rDi collapseover shallow magrnachanbers, long tenn slability criteria indicate that the equatorial crusl about 7 l0 kn behw lhe snrface. shouldbe dcsiccatedtosignificaDt depths [361. Mouginis Mark Tf this intcerelation of the origh of LheMedusae Fossae [35] thcrcfore considercd thc mosl likely explanxdons for the fon11ationis broadly conect, tbcn one can speculate'neaning channels either to be the melting of ice lenses deePer than - l lully over a 2 billion year hislory of the area.The ignimbritcs kn by an igneous inrusioo, or the tectonic release of the rvere formed as a result of rcpeared, large-volume. eruptions, conlents of a deep groundwater systen vi,t frtrlls caused by the inteffpersed in time by ihe incursion of lava flows liom the great mass of Olympus Mons loadirg the surroundingplai's plains to the nolth. andTharsis and Olympus Mons to thc cast. Both of these models imply that themrodynamically useful Thesedifferent sty lcs of volcanismwould havercflected diftef- fluids may slill be presentat rclrtivcly shalow depthsbenealh ing miner{bgic.tl xnd voiatile conpos;tn)nsof the magma at thesepl.ti!s. Obviously.ifthe fldrdinghas beenassociated with source thxl rising benealhsoudrem Amazonis Fesumably be' high hcalflow thenthe potentialof thesiie as a geothemal field ing norc silicic,.nd hencemore viscousand liable to explosive would be much enhanced.Pcrhaps if there are any hypcthermal Plate3. Geologically rccent flooding on Mtus [351. Wlter appedsto havcbee! releasedlilm this arcuaLeeubctr located15 km to rhe e.sr ol rhe OtympusMons escarpment,poinrnrg to an inLeracrionberween groLndwaler and a subnerycd heat $urce. viking liane: 468553:.t 16"N,129'W: image widLh: 23lon: resolulion: 24 D/pi\el. Photouunea of NSSDC

areas remaning on Mars ir rnight be here ar the foor of thc shed dark streaksdownlill and into adjacenrgutlies. grcatestvolcano in the Solar System. That thesc Coprates patches are exremely young seems indicatedby a number of criieria: 3.5 Valles Marineris (i) their morphological crispness and feathery oxtlines; On the Eanh, volcadsm is commonly associated wiih rcctonic rifting, bui i. the large Valles Marineris rifl systemon M:rs (2) thcir superposinonon Upper Amazonian landslide obvious volcanic featuressuch as shieldsand lava flows ,re deposits(Unir As); and absent. Neverthelcss, a detailed survey of the inrerior deposits of the central Valles Marineris area by Luchifta ha1 revealed a (3) their dark colour. The laner is signifrcanr because dark variely of more subtle fearwes oI possible votcaDic origin- igneous deposits on MaIS would generally be expected some or which may havc been lbmed recently t37,381.The to lighten with age due ro oxidadve weathering and most compclling evidencecones in rhe fonn of dark patches mantling with dust- By estimaling the rares of thcse which can bcenseen ftom high resoluiion\4k;ng imagesto line processesbeiween wide bonnds, Luch;tta calcular,ed up with iectonic siructurcs.Those in Ophir and Candor Chasmata that thesedeposits could have been fomed as recently are found mainly at the base of scarps found on interior layered as 30 yearsro 1.5 million yeaff. depos;ts, whercas those in rhe norLh of Coprares Chr,sma occxr adjacent to alrd along the fautrs ar the base of rough wals. If volcanisn associated with the Valles Marineris is as young Thesc dark markings in Coprates Chasma point most persua- as thistheniiis unlikely to be totally exti.ct andlargefluxes of sively to a recent volcanic ofigin, exlending disconrinuousty mxgmatic heai may slill be leaking ont iion beneath the ca.yon along iis nothern boundary fault line lbr 200 kn. The fhcr ttrat floors- The presence of adequate $oundwater for geothemal theyjuxtapose so well with such large-scalcstrucrures argues power putposes cannol be assessed lrom the curenr dara, agninsi thei. being rhe resulL of aeolian deposirion. Moreover, howevcrthe explosivenature ofrhe local etuptionsimplies thal they appem not to be derived fron exposed volcanic silts or the magma at least was, or perhaps sn is, 'or lacking in dykcs eroded by the wind as they lic on top of lardslide deposirs volatiles. which would have buried such ouroops on the canyon floor (see Plale 4). The Fesence of pyroclasric deposits seems b exptain 3.6 PossilrleCrJptovolcaniclncations all the fcltures of the patches rhe besr their wispy appcarnnce inplies explosive jefting of material their colou|s are snnilar to It has been noted above thar 4.5 million km, of Mars is sljghdy wcatheredbasalt powder: and iheir low albedocsae geologically characierised ds Upper Ainazonian in age and of simila! to ihoseof pyroclasticmantles idennficd on the Moon. volcanic or fluvial odgin- These sudaces rcpresent rhe overl The ertusion,,f l0rr ma) J1".,be indrcarrdin \ome:rir.s evidence of receft anomalous heat flow Additionai sigrifi- beneaththose patchesvisible h thc Cha Ila wall which have cancecan be asc.ibedby their gcogr.phic location allofrhem The Ulilitt afceotltmal Erery! .n Murs

The Cotrates Chasma floor !! l2'S. 67'W, showing Lhe7 km high lotlh wall ol the chasma ar the top of lhe inage dd a lddslide deposir exteding lrom $e sonth. A batrd of wbpy darl latches are s@n ruMiry along the laull trace at thc base ol lhe wall ed may represenl explosile volcdic vents of sent ongin 1381. Vitilg frme: 80A01: imase widlh: 80 km. Phota touflesJ of NSSDC

cluster into one distinct province betwe€n roughly 20 220'W association of martian grabens with recent volcanic and50'N l5'S, m expanseof about40 million km?,equivaleDt xclivity has already bcen meniioned in the contexls oi to 287, of the 1o1alplanetary area. Am.tzonian volcanism, that the Vales Marinens and Cerberus Rupes- Howevet lhe younger th:n about two billion yeirs, seems to have concen ATIFP contains many more such tectonic stuctures, on trnLed here, perhaps as the surfacc manifestation ol long term a variety of scales, which could have bcen accompaied heat loss ftom a mantle "supcrylume" beneadr.The facl thrt this with cryplovolcanism. Another complex gaben system Foposed "anomalous hcat flow province" (AHFP) seemsio that might merit investigation for heat flow anonalies have been active for sr long, and inlo geologically rccenftimes, dre the Ceraunius Fossae,which appearto have fractured makes it perhaps unlikely that all magmatism has ceased. Noachian rocks in Lowcr Amazonian time. but which Moreovcr the overt signs of volcanism that are visible .rlr?osr also appear to havc had something to do wilh mnch later c.ftainly understatz the activitJ that has actualb occumd ^s fluvial activ;ty f351. considerably more magma than the amount rcleased to the surface could have been intruded at sha ow depths. Such (2) Turbidite Arcas. Cryptovolcanism is parricularly "cryptovolcanism" shaUow voicadc iniusion withoul eruP_ favourcd under areas bnried beneath thick plastic tion-is common on the Earth and it is notewothy Lhalwhilst all sediments,such as silts depositeduDderwater This is comnercial geothcmal fields are associatcdwith regions of because clays may flo\t into lault spaces in the hard present day, or Miocene to Quaternary volcxnic activity (< 20 bedrock beneath, sealing ihem against the rise ol Myrold), not all ffe sitedclose to volcanoes[15].lnthct, some upwelling magma. Rising intrusions with insnllicient of the most productivegeothemal iields, suchas Lardercllo in energy therefore nay come to rest at the soll rock/hard Ikly and the Geysers in Cllifomia. draw their power from rock boundry. This might pafticularly apply on Mars hidden magmatic sources- where thc lower gravity resulLsin lesser buoyancy Whilst cryptovoldlism could occur anywhereoD Mars, itis most reasonableto invoke iis possibility within thc proposed aDonalous heat flow province. However. cxcept where resolv- Bolh Larderelo and the Geysers arc exanples oI turbidiG- able outburst flood features have been created by magmatic arca geotheflnal fields and may have their manian analogues. 14) irru.i"n !'ucl- a. rc!,re$edIn Secri,'n vil;ng rmagescrn SubsLantial areas of the northe plains are inchded in the provide no direci evidence of this process. Howevea certain AHFP, as well as Chryse Planitia witb its promineni Hesperian geological envionments on the Eanh can be indicative oI ouiflow channels- l-arge thicknesses of water-lain sediments cryptovolcadsn [5] and have analosues within the AHFP on may have been deposiied in these localities-especidlly if the Mars. hypothesis of the past existence of a mariian Boreal Ocean is coffect [39]. That Upper A mzonian voicanism has occurcd on (1) RiJlvalleysandlaryecftt ?rr.Thesetcclonicstructures .he nothcm lowlands is indicaied by the uncratered lava flows are causcd by tension and thinDing wiihin the ffust and small volcanic cones inaged in Aradia md Acidalia resulting in the downfaulting of l,rge longitudinal blocls. Planitia(UnitsAar andAa).The occunenceofcryptovolcanism, Often, the relative pressure rclease that rcsults at depth sealed beneath the sediments of large adjacent areas and unde results in a rise of magma in compeffation. The possible tectedby the Vikings, is a distinct possibility. 3.7 HJperthermal Localities on Marsi channels.the martianregolith is thoughr1(] be borh porcusard An Opposing Conclusion pemeable (hishly so in sone arcas l40l) rnd ro possessa groundwatersystem. possibly ofglobal exrent[241. Aquifers on The fact thal volcanic xnd fluviai activiry on Mars has,for the Mar! are automrlically sealedby Lhecryospherc (the upper past iicap two billion ye s. beenla.gely coDfinedto oneqnadrant of latcrs of cnrsi below fieezing) and so a rock:i of lorls thc llanet, and the faci ihar it has coniinucd inro the rccent would be ubiquitousthat iight acrrally collflne groundwaterat geolog'cpast, strongly suggeslsthat rhe occunenceofprescnt low elcvationsunder aflcsianpressure. The gfearesrdoubt ties day hypethermal areascannoi bc ruled ourby arEurncrs oasco in the presenceof geothcmai hearanomalies bur ir has been on mern gldrdlheat flow calcularions.The escapcof heatfion] noted abovethat thesetoo nay srill penist in sone places_ the inkrnrf of Mars has evideody bccn vefy parclryover mosr Of coursc,there wil be sysrematicdillereDces between rhc of the phDet's history and it seemsnorc likely than not thar this versionoffig. I that appliesro the Earth and thc one apptying stateof aflairs should slill pedain ar Fescnt. Thus, whitst the b Marc. The lowcr temperatureofthe sudaceon MaN (>60 K decp frozen Mnrs paradigmmry be valid at gtobal resolution. cotdef) means tha1. for a given surf.tce hear flow, suitable ir may heak down with thc gaining of dctail ar regionat. or rcservoifsare going ro be at grealerdeprhs: bore holes must iherciore be systenalic,tlly deeperand more expensivcand a In the opinion ofthis aurhor our presenigeobgic datalrlc greaterquantity ofnseful heatwill be lost during ftnid cxtrac onbalance supportiveofrhepossibilityoflocaringhyperrhernxl rion by conduction into surfounding cold rock. However 4 areason Mars, mther than the opposirenoiion of a uniformly nidgaling fnctor here is that krw temperaturc fluids on Mars are cold crust. h is thercfore celtainly worth invesrigatingMars thermodynxtnicallymoreusefula supplyofwarc. arrhe equiva- fudher with the specilic inrendon of prospecting usetut lent of waste-watertemperarures on rhe Effth (< 30 40ocl geothemal resources. would snll be prcciouson Mars. The lessermartian graviry has a numberofconsequences. Pressure is lessar a given depthand 4. A MODEL OIiA GEOTHERMAL FIELD so for a given fluid temperatwe. there is a greater lltelihood of ON MARS a siluation where vapour pressure is found to exceed the hydrostaticpressure: all orhcrthings being cqual,sream-doini So as to characterisethe nature()1-the geothernal resowces wc nated i:iscrvoiN (prized lbr clectriciry generation) may thus bc ]rnghl cxpect to llnd on Mars, a sinple model of a marrian relalively more conmon. Punping frcn warer dominatedres, geothermrl field is prcsented in this Section. Thc general ervoirswill requireless energy, but sincepressure gradienrs and featwes of such a field and the method of irs exptoitation a bnoyancytbces will be 1ess,fluid circnlationand hcnce narural concepi known as the "geothennal loop" are sketched fig t rcchargeby intlow from adjacenrffeas will be morc sluggish. Essentially, heat conducied from a high-energy source benearh Rc injection of walers will rhus be even more impo ant on misesthe temperature of $oundwater within aconfinedaquiiei Mars, not jtrst to maintain its circularion.but to conserveils Thesc hot fluids crn be rappedby drilling bore hotesdown ro supply as recharse of the aquifef as a whole thrcugh precipila their location and. dcpending on such facrors as lne rcservof tion over.reas whereit outcropson rhe surlacewill nor occur temperature. pressure and pemeabilirl, the flow obrained at the Some simplc mathematicalmodell;ng can provide apprcxi- js well head ei$er in the form of hot warcr superhealed steam, mrte esiimales oi the potential heat rcsources availablc benearh or a two-phase mixture. At the surface, thc enrhalpy of these thc martian surface and the depths where useful fluids night be fluids can eilhef be usedfbr dn ed heatingor the generaft of iocated.The magnitudeof the verrical heat flow ./ W m, js electriciq,(discussed in Seclbn 5). The cooled rcsidnalwater relaled to the crusral lemperarue gradie.t d/du K mr by is thcn injected back inio rhe aquifer ro be re heated and re Fonrier's law of heat conduction;

This descripnonis somewharslylised and d,r,c 4rc lrary , .!t variants of it on the Earth. lt conld also sefre however as a (t) protoiype of a geothermal field on Mars &.' its main characier- da istics could be duplicated there. Jrdging by such geo'norpholog;cilfeatures as the va1leynelworks and outftow wherei;s the hermalconducriviry ofthe rocksin W m., K,_

From this it foilows that, assuminga constant temperarure gadient,the depdr. benealh the sudace to a giventemperature

. t(.)-t c)

where r. is the sudacetemperature. ln order to proceedfrom herc, it must be noredrhat as on E8nh the characler of rocks will vxry from place to place. lnfluencingpropenies such as their specifichear and column mass.However since order ofmagnilude accwacy suifices fo. lhis study,it is sufiicient to assumehomogeneiry and, over ihe depthsbeing considered,a constanrtempcraturc gradienr. The consiantschosen to delinethis mean crusral rock arc a densily r ' Flg. I TIE GeoLhemalLoo!. Exploitaiionof georhc,',Eerergy Dy of2700kgm anda specficheat capacity o1839 J kg K,. From the exttuctior of heaLedgroundwater and irs rcinjectun irto tte theseone can udculate [4i ] thalthe cooling of I kmr of fockby aquifer (after Rci t.+ll). I K releasesheat energy of 2265 TJ. We thus define a quantiry

4t2 Ttt! Utilir\ oJ G..thenndi En?tlt .n Mdts knowD as Lhcs7,".,,f. .rerldl ll?4r to bc: q = 2265 TJ km' K whercthe faclor of l/1000 in Eq- 4 converts.fr-,J lrom m 1() or 7l.83 MWyt km I K'. The energy released by a givcn volnme of rock. cooled by a The resticted heat rcsource O. thus gjves thc maximum given increment. will be thc sane anywhere on Mars. Ho zr"/ amouni of useful heatrvailable beneaththe sn ace.However. it ^ alranlugeous 10 ria* htpetThertndlareas becausethef this upper lirnit would never be appronched in practice. A cotrtuina much greater tatzl heat vhich ttill be a.cessihleat leasible heat mining project on Ma|s would aim to extact a shallo\rer tupths. This rotal cnstal heat can bc gauged by quutity of heat perhaps equivalent to 0. muhiplied by a few plotting a graph of lcmperatureveNus depth 1411.The area deg.ees not by auniformcooling oflhe c stalcolumn. but by undef the cuNe (cxpressedin k]n K) nNltiplicd by O" gives the cu"li ,g r\' ,,quilerro.L. ny J Erra..rmrgrurr de tdal crusLalheal abovesudace tempcrrlut rvailable beneath In order lo complete the model, values of its "coDstants" each knrz of land suface. However nol all this heat will be musl be assigned.ln line with our pnvious xssunplion of a(ainableor uselul for geothcmal power Foduction. Rocksof crustal honogeneity. wc chose a unifom vrlue of ihemal = Kr and perfom a cetain minimum temperaturc1,,,, nust be reachedand ther€ conductivity equal to t 2 W lr1r [22] iilr DerdeornLor.- he).n,i ler. rruningi' i , .rJr calcrlations appropriatclo tropical locations(- 50E the plan- "htr\ iical. This Limits the heat available lbr practical purposes to thc elary arca) wherc ncan sudace temFrature is r, = 220 K Assuming a watcl snpply at a tenperature as low as 30"C is Temperalure.t(K) economicallyuselol on Mars and.assuming tunher thatthere js a 30"C drop nr |empelalure betwccn reservoir and well heird, then thc ninimunl useful rock lemperature that must bc rerched is t = 60'C. Geothemal fluids ftom conrmercialtields on the Eari'il are usLrally obtaiDcd liom deptlN of0.5 2 km; on colder Mffs thoush. such rescNoirs will inevitably bc deeperdown Since the currenr state of the alt in borchole drillhg alows penetatio! to l0 km ibr vefical wells rnd 6 km tbr deviated wel1s[41], .-^ is selto eachofthesc depthsand calculations of O are done for both. A depth linrit i. anothersense is approxi = I 0 km, in so lar asporosity this far '-E mrtely rcpresenledby .,,,- dowD is expectedto tend to zero due to self comprction of the nrcgaregolithby the wcighl of rock above 122,241- Applicalion of this model to what might bc regardedas "iverage Ma|s crust" is tuslructive. Here, thc heat flow is set &) r the global mcan of ./ = 0 03 W nr dcnled by others116l The temperaturc gradient is therefore.l/d. = 15 K km ' and the dcpth ofthe0'C isothem is - 3.5 km andthat of lher,,,,, 'sothenn - 7.5 km. Tempemturesoniy rise to 100'C ai l0 km deplh. The rcsticledheatresourceabove 6kmis zero.butthd above10km is - 105TJ kln ?. equivaleni to - 3000 Mwyt km']. Perhaps surprisingly tlrefeforc, wa rm wdtet bet een 6a" ' l a''C night l' alurn oa)a l,"rr b"re rh th, Jrttoaltuu;.' e\.r ",ritabl. i:l |h( pessinistt deepJnzen llars paradism is ttu. How.vct, depthsrhat lraveto be drilled to i. order io obtain thesefluids aregreat and into stratawhere porosity is tendingto a mini'num. Fig. 2 Definltion of the RcsirictedHeaL llesource: thai hcat contanred A low cnthalpygeothermal rescrvoir wilh thesecha.actcnsdcs in o,,. abo..I.cr" '. r'ul r peflr.re ,o'lu$, r, ,.d' n on thc Earth would be tdally uneconomic.Whether it wouldbe (z,h\). attainablcde n so on M.rs, where any suppl) of liquid watef would be higNy pnzed, is debateablebowever it would cert,tinly be less than areadcfincd by the shadedzone in fig. 2. ideal. \4"r. o\ r\lanr mdgmrli.m This heal thar per km? of sudacc, at altainable depths. above Hlper.hern,:'l!reb. un herleJ sincc such horspotswill be a minimui usetul temperahur:, is called t}'e restticted heat are tlus exhemcly desnabb but 1o add a time dlmensrcnto '"ro&,t€ (O.TJ km') and is here calculatedas: geologically transient,it is lelpfnl the modcl. The declirc of heat flow fbllowing $e wannrg of magmaticaclivily from its pcltkis assumedto obey thc follow- /- ,/, n/r' t ' l nrg crnpirical fbmula: u tt r..tt 2 q@=+ where zft-,^) is thc depih at which the nrinifrrm useful "' 'lr !o.) temperaturc is cncouniered and tfr-,,) is the tenperarure at the naximum auainable depth. These are obtained trom the folow whcreTis lime in millioD yearunits and 40 is the pcak heal flow For ocearic crust on the Earth. the above rclation is ? observedto hold for 7= 3 120Myr with q, = 0-473w m [42]. .(1,,,,)=ftffi How applicable such a relarion would be for Mars is unknown as the cooling of magma bodies is a complex issue and lhe geologic selting would be differenir vxlucs ol ./(Z) denved herc t(..,")=T+t. (5) theretbrc should be treated with ca tion. The restrltsoi rnodellingthe characteristicsand evolutionof TABLI 3. Resubs af th. M.tftian Geothernal Fietd Mo.let

Arei lipe Waning hotspot

Age: r/ Myr > 150 0l l0 20 50 r) Heal l.row: ( / (Wo 0.01 0_366 0.1t6 0.082 0.052

l5 l8l lJ2 5rJ )(t (dt/d.) | lK knl) Deplhof 0"Cisorhenn: 3.53 0.29 0.65 0.91 1.29 2.O4

Depth of 60'C isoLhem: 7.51 4.62 l.llJ r.95 2.16

Depth ot 100'C nodrem: -t0 t.t37 2.64 t.7:l 5.lJti

DepLhof 200'C isoLhenn: N/A l.3ll 1.09 6.t1 9./3

Depthol300"C isothem: r_92 4.30 6.09 8.61 N/A ..(573K) kn __T__l

37.C(3 r0K) 1045"C(r3l8K) 439"C(712K) 193.C(.166K) (3?0K) >r227.C(>1500K) 52?"C(800K) _97C __.1 20r.c (480K) Kcrtrctcd hcar rcsou(e:l L'a:!y.r I _rY"oa.l Qk,-= 6 kn)

Q,(.,^= to k,l

\\ Fii ts, li 170 times lrcater than the the sudace, thcrmal springs emc.ge in sone places wirh sufil O, calcnlaledfor avefageMars crust. cie.t energyto causeoutburst tlooding, funher ground subsid- The qualiiy ofa geothermalfleld willdecline over rhc next ence,and resurtacingwith dlluvial sediments.Eveniualty dnd lew million yearsfblkrwi.g the cessxtionof m.Lgmaricactiliry. assislcd by the ple cxisting fauiring basalr enprs through Tbc modelcalculations for 5 l0 Mla show thatby rhis rime thc fssnres, in a seriesof pulses and in a number of tocations. coDlcntsofthe nagna chaDberhavesolidificd. However,hear buryng Lhcpreviouslandscape under several hundrcd melrcs of flows rcm.rinquite high a1> 0.1 W m1 and rhcrmal gradienrs (whilsl 100 degreesless than prcviously) are still > 50 K km'. The situationbeneath srch an eruptive cenrre.ilnmedialety The deptlr of the cryospherercnranri lessthan a kikrmetreand following fic climax ofits aclivity (I= I Myr) is illustratedin so the addilioDaldrilling reqniredto rcachusetul rempc.atures. fig.3b dnd quantifiediu Table3, cohrmn2. Themofofabasatr when compired 10e.tuivalent operalbns on the Earth,is not all Th. L:tilit\ .fG.otlt nnal Etltt(\.n llth

deeFr than - .1..1bn. Whilsl hert minilg is drcorcLically possiblefrorn srlchsjies l(' cle trcm alerrge Nlarsc sI".as a. q:0.03 w,m2 22o K nored abole) it n ) be thrt conlettional gcolhenxrl poiver production would be nnpracticalor trnccononic. especiallyif ihe charrcle.isLicsof aquifer xt thc tlcprh\ requifed do not I pefDit rn {dcquale raie of fltrid rcco\ctJ. However. amificirl I medr)dsol rcse oir slinnLlrtioD.such rs proposedfof c\ploil- ing "hot dr! rock on E{fih. night signiicantl,'"cnh{nce the i,lir I ltililr- ol mildlt h)penhernal arers on Nlafs (scc Scction 6). $re ight rheref(ncconclude tionr this smDle nodel of n nanian geothem i ficld lhai\'olc an ic of cr) pto\ olc]rnichoistols less than ten nrillion )ears old offcr a good polential fof lhc ,:1.373K:rtl produclrono l geothernalpower. I l T.r akae/r/.l20lr.e.onccl andthefrrc ol cim!i\,e rnagnutisniis ore eve|t evef] - I0.00t) yexN thcDo e night expecton the order of - l0'/10r = liltl0 \uch holspoB on l\'lars.Allo\nrg lor repeatedc.uplions nighi b. 3f,":'j'"lt11lH rcduce dis number by r lactor of perhxps l0 10{1.includrng crypLorrolcaricsources Lhough\1'ould itc.easc it.Whailiacdon of these hotspt\ N.,uld also occur in conbination wrth an adequrteg(runcl n'iter sutfl) is lncertain. bul wherc coDdi- tior! are stritrble.borehole! ofscvcral kilolllene! depth $ould acces\cDcrgl resoufcesofthousx ds ofnegnwrtt )cars (lher_ inrl). belorr eachsqMr. kilonere ofsu,face. in thc 1onl ofhol lluids. Suchfluids coul.l be 'rsedby a grownrg n1.1ttianciljlisr tion in a $ide larict) of $al,s.

5. I]SES OF GEOTHER\IAL ENERGY ON NTARS

l)eteDdins or their geologicrl seltirg. geolhemal lluids are obtliDrblc aI a ll1ngeof ternpcrrlurcs and flos rat.s. Rese oirs 'C arc citegorisedas heing o1/otr d,//r.l/pr if bcLwee 30 - 120 .\ii high enthdb.i if abole Ll3 111. Thc tbrmer sufply waln !\'atefsffom r\hich lhe sensiblehcaL c.rr be recolered rDd the lallef fr)liclc hot pressLniscd\lale]. I$o phasemr\Lurel. or st--r.r tlhich yield latent hcrt in rddition. LlDdesirablethrld chrracrcristicscxn in.ltrclca rich conlenl of dissrhed solids or non condensiblegalcs lhrt mighi causesctl g .. coflonon ot cquipmeni.A \arict), oI producliontechnologies and utilisatioD smlegies are thercforercquired f(n thc oflilnun] recove|! of heat.The eDgincenrgdetails ofthcsc arelengthy atd il \uifices here to arguc in general!em\ fot dre side polent;rl uLilill of geothcnnrl po$'er or NIaN. Suchpo'\'er is obtnned lior geofietmxlfluid\ oirherby e dir..r u!e offieir heatconlent, oribe l/r./rt?.rLransfbmtion of Flg 3 SclDndticoltlre NIdLiangeothcrnll li€ldn del. rhrr hearinto ele.ni.ity On rheEafth, gcoeleclricgenerr . $crage \h.trdn jdsl l.1,ll. pefhafs of disproportionateimporlaLrce because dircct b. nnmedirlellfollo{in-q thc peakrLiviLI of flood lolcdri$ lior is 1{)lhc \iciriiy ofthe rei)urc. c Tl,e lield xfter- 20 Nlrr o declnrelsee alsoTahlc il heata|plicntions arc corsirained sneiiself. $hef.ds clecmcity is.eadil) u.rLlsporlableo!er l{rge distances.On the other hnnd. dBjcl nse benelir! ftrrm a nrrch - ' that Sfea!.Wrm t! rlcr \ ould be encountercdbck^\' 1.:l 2 krn greatercon\eAion etljcjency (lhe rl1ro betNeenu\cful e11erg) and hottef fluid\ cl.eper than - 2 3 km. Resfficted hea! outFrtand thelhermal eDery)r releasedlionl thecoolcd geofluid) fesdrces. \!hilst sc\er tines lessthrD iDitixllt. are siill lxrge: \rhich nr,\, be up !o 90'l lbr afplicdions such $ heating.btri bet'lccn115,000 - 219.000NI\l krnrclo\lr io lrikm.'11 6a) lcssthan 209n fof geoclccldcpower. l\4orcolcr. o colderNlab. timcs grcalerthrn averagecrusLal 0.. Such area\ the.eforcslill lo\\'enfi:rlpl fluids thal can onl,'" bc ciploited difecrl] x.c hrvc good polential fof trcducmg geolhermalpower nilrinsically nx)fc uselul amdthe sitc specilic natufe of difcct Alier 2il Nl)r ofdeclire, thc nlodel predich r hyFflhcnnal heat applicatioDsrnay be lessofr.onsrrairt sincedrefe {fc no afe! with a hear flo\r of - 0.08 \Y nlr similu to Llrccrustal pre'existing ccnues of civilisatidr. Settlemenison Mrrs lrrc alerage for the Eanh. Eren though 0, l.=/, t,l is - 2.1times likely to be tbunded {dtaccnt io that plnnefs ich.d natural ihe ma|tiancnLstal .]\crrge. usetultemt.ratures arc no$,lound fesourccsgeothefmnl relotrires included. quire deef dowD rncl isoiheflns afe strcrdtug onl wilh the feduclion in tcrnpcnture gradient (scc fig. 3c). \hrn waters 5.1 lndircctUses:(;eoelectricity would bc obranredAom between - 2.1 - 3.7 krn and highcr cDth.rlp)fluids (ifrecover{ble at rll) trrm considefabl,""grcaLcr -,\rcbnsi and reliablcsonrce ofeleclricit) \ill be requiredby {ll dcplhs.Bj 50 !1,\'r.heat Ilow has[1]len loless that thetcrc\trial nariian settlemcnls.Such power crn be generatedfrom high lrlerage and.ivhilst stillclc\.tled in marlian temrs,0.is only _ enlhalfl gcotheinal fluids eirhcr br- drivirg a lubnrc directly 9 iimes gfeater and usclul temperantresh{!c to be sought with geothonnal steanr. or with ar organic working lluid vapoised vla heat exchangen by hot pressudsedgeothernat sragemay represeni' 80q. of thatextraclcd and in orderLo avoid js waiels (see fig. 4) I44 461. The lbmer method the mon a complelc waste of the hcat it stil conrains.and ro maintain commonl] praciised,bu1 i! nost suitablcforusewith gcothermal reseIvolr prcssure! it is nsually fe-iDjecred rnto rts sour.cc - fl uids abovc 150'C; the latter"binary cycle" is morc ctricienr aquiter.The simpleir way of dischargingthe stean from rhe at generatingpower at lower temperaru.es.Typically. one trrbine is io exhaustit to the atnosphere On theEarh however, production - well nust be drilled per 10 MWe outpui 1brpower this is inefiicienl rs the oudlow has to push againsta I b:r - - stxLionsrated between I 100MWe. Assuning a demandof ahospheric pressurcand therefbreLhc hear&op behw I 00 .C 20 kwe/persono. Mdrs (discxssedin Section2). one suchptani is wasted.(ExhaxsLlng to atmospherccycles are also contra, would provide for the ne€dsof 50 5000 people. indicated at ratings above - 5 MWe on vafious environmertal Geothefmal wells that producc dry steam are rhe most grounds.)The soiution is kr exhausiinro a corueDserar suD- convenientlbr generrtingelecriciry asthe vapourcanbe piped atmosphcncpressue (e.g. 75 mbaf ar 45 'C) which is less through a demisrer saaight to the rurbine inlet. The more wastelirlofenLhalpy. but is boughtat theprice ofmore comptcx commonly occuring hor pressurised water or rwo phase flows and expenslvcequipment. however rcquire steamlo be separatedbefore utilisatnrn. The Solrleofthc envifonmentalparaneters of Ma|s are actua y fluid is piped to a sepiralor vessel wherc it is flashcd by a better suhe.l ta gcoelecttic power generaiion than thejr terfes- pressurc releasc that optimiscs ihe flow rute and enlhatpy trill equivalents.One of theseis thc hck of liee oxygen which contentof lhe outputsteam.Ifthe thermodynitmicsare tavou. pro'noies corroslon, especially in the pfesenceof exlolved ablc, the residual hot water is ther pui through turiher flash gascssuch as H,S. The most obvious narian advantagehow- cycles,edch one producing a lowcr pressuresrerm that is also ever is the - 6 mbar atnrosphericpressure equivatenr to the 'c. conveyedto the turbine.The waierrcsulting fioln the lasrltash vapourprcrsure of water a1- 0 Thus, simpleexhausring to atnosphcreturbines can exploii an exaa 100oc heai drop and, lince ihe cntire cycle is at supra armosphcricpressure, no condensingequipment is required. Environmcntal problems with noiseand malodorcus gases erc are also likely to be lessof a fxctor. although the poteniial nuisance of ddfts of snow accumulating around the power slation would have ro be assessed.Asit coolstrom 0'C to -60'C, this snow would waste a substantialamount ofheat (includinghtcnt heatoftusion).In order to avoid ihis, ii might be bestro mix rhe turbine exhaust with an)r wann water lefl ovef from flash separation befbre rc

Only a deiailed assessmenrwill clarify a[ the engineering challengesinvolved i. constucting and nlndng a martian geoelectric power plant. Whilsr rhere are likely io be some obslaclesunique to that planer to be overconle, the basic technotogy is alrcady rried, tested and understood. civen thar geothemal fields are present,Mafs may actually be a befter place fbr geoelectric power generarion than thc Earth.

5.2 Direct Uses

Direct geothemal heating is one ofthe cheapesrforms of energy and has a long iladition. Bothhigh and low eDthalpyfluids can potemially selve a wide range ofapplicarions I47 491 and sone of iheseare l;sted againsl the necessaryfluid remperatulein the Litural diasram showr'infig- 5 [48].lt shonldbe notedthan an analogous diagram for Mars will have much in common but will notbeidentical. Soneoftheenaies in fig.5 will not be felevanr to Mars: similarly, there will be uses found for direcl geothermal heating on Mars thai are not needed on $e Earth. High enthalpy fluids .re applicable to indusidal processes that benefit from a supply of saturated steam, such as in the manufacture of a wide variety of chemicals (including ethanol, //-\condelM hydrogen peroxide and vitamiD C). paper, rextiles and plasrics. Orher potential uses include food processing, wxter distila.ion, \v/ co.lant salt production, nineral extraction, air condirioning. and ce- menr seiting 1481. Although applicable however geothermal steam has rarely been adopted for lhese processes in pracdse. sirnply because of tbe jnconvenlence of co locaring indusrries at geothermal sites and the fact ihar gcorhermal technological solutionsarc lessfamilirlthan thoseinvolving heatingby fuel Fig. 4 Thee methodsof geDemtinggeclecrrtc power oil or electricity. On Mars however. civilisation wil be making a. Exhaustinggeothemal stem ,;a a turbine 10 rhe atnosphere. a fresh st:n in the absenceof fossil fuels and pre existing cenrres b. Sepdatingsteam from a iwo phaseflow wirh mdnple flash cycles_ of power demand. lndustries will have to be configured ro suir c. The Bindy Cyclc using Seothemal warer to heat a secondary the natwal rcsources available. If geothermal steam ;s discov- working fluid. €red to be one of those natural resources, rhen ir undoubtedly nE thihr .l A.oth.tual h)tt&! d Mnr:

.C Geothemral heating is used on ,t sn ller scale thoughout Iceland as descdbed in this fascin.tting paragraph from a 200 relelanl prper 190 l47l: 180 EvaloEtionof nighlyco.centated sohtior "Sinte aboltt 19lA eLenentzry and secondaDi boa ing E Rehge.atbn by atnNoria absorytun Wood procesing ln papc. manrtu.Lure schaolsin the runI arcasol IceLanl havewhenerer possible 170 Heavy water vla hydrogen sulphideproces bee sitetlatIocationswheft geothernal enery! is druildbLe. Drylng of diatonaceoustiarlh In thete centrcs the school buil.dings and liNins .tua ersfol i60 Dryingof llsn meal th( pupils and stall are geathernally heatz.l. Ther are also as Dryi.g of rnnber a tuIc equipped \|ith a strinnins pool, and d ft seU-sttpplying 150- Altrminavia Bayefsproccss with w getdblls ( tonatoct, .utunbers, caulil.,wers erc.) gnwn in theit own hot houses.Therc are now nary suth 140 Dryiry flm lrodtuLs at high rales sthaoh h variout pdrts .tf the nuntry, and quite ojlen thq, Cdnllg of tiod dre used a! toutist hotek durins the s runet holidaJs. 130 lvapo.atbn in sugarreflling Quile .n new sewtce Extracd)nof sallsby evapomtionand crystxllization ol1en these ca tret hde lome.l the nuclei Fiesh water by distilhtion n tunities in the rural iftas. 120 Most multl cllirct evaprdlion. Concetrtratiotrof sailnc The an^k)gybetween this descr;ptionolisolated conmuni ties in Iccland and the vishn of selllementson Mars which I l0 - Drying xnd cnrlng of llght agg.esare.emenl sl.bs must be similarly self sufficjenl and the nuclei of grcwLh to come is nost striking. Evidenily, lhe experienceoi settling 100 Dryingof orylnicnatendls, seeseeds. gras, vegetablcs Iceland could have significant relevanceto that ol settling certain areas of MaA. A large proponion of thc basic energy Wlslri,rgand drying of wool needsfor a mxnian societywill be for herting. On most of the 90 Dryirg ofstockfish with elcctdcalpower, but where lnrensede iciue opcmnons planetrhis wiU haveto be done geo$e|m{l fhrids .re presenta direct heatingoption is at hand 80 Spaceheating (buildnrgs dnd sreenhouse9 that mry bc simpler moreelTi.ientand inexpensive (see fig. 6). 70 Rcfiigeratidr(lower terperaiurc li'nit) Thc facl that low entftalpy rcscrvoirs are likely to be much more E conmon on Ma$ thaDthcir high enthalpycounte'IJ,tns does not 60 Animal hnsbardry conpronise the feasibility of exploiting georhenn.tl energy. GreenlDnsesby combinedsr)ace and hotbedheatjlg The narurally occunnrg energy distribution nray adnirably 50 Mushmo'ngn)wing malch t|e dcnrandsof the mafian heatnarket- ,10 - Soil wannirg

30 Swinnlng pools. bn egradrtio!, fermentltlons Wannwabr nt yeir rould tnhirg in cold clnnaGs

20 Halchingof flsL Fishlami,rg

'llre Iie. 5 I -ind4l Dixgrnn. A tenperaturc spcnum ol polenlial dlrect gsnhemal healingapplications (atcr Ref la8l).

increasesthe techrological oplions lbr maitian indushf aid opponunitiesfor growth. Low enthalpy gcothermalsystems that yiekl hor wrler .,€ relatively more conrnon lh:ln steamsystems aDd alc cxpected to be cvcn ore so on Mars because of gcncrally lower heat flows and surface temperatures.Such rcsourcesare of little practic.l n se fbr geoelectrical gencration and are best suited fol direct applications.Foftunately. such applicationsabound (see 1lg.5) andcertain peoples suclr as theJapaneseandMaorihnve taken advantage of hot springs fbr bathing and cooking sincc pre-history.Today, well known examplesof thedirect use of krw enthalpy geothemal encrgy are district heating in Iceland .tnd greenhouseheating in Hongary [47.49]. Most imprcssively,the Fig. 6 Dircd useof geothernialcnergy nr baseloadhabitat heating. c;ty of Rcykiavft is alnost entirely heatedby - 435 MWi of Peakloads night be hadlcd by elecbically poweredluxiliary heate^ geothermal waten pumped from three neffby fields at a rate ol - 1970kg s I. The sysremprov;des aboul 1.5 mr of piping hot waierlor eachof 120.000peopleperdayand is designedto meet 5.3 Cascading Usc ihe heatiry load at 10 "C outside iemperalwe. Watef at an initial mean temperatrre oI 95'C is disrributed and passes In order fbr martian pionccrs ro plosper, it may be crucial ibr ttuough radiators and he.tlcd floo and is of such good chenricxl ihem lo rnaximise thc efficienl use of any given resource. quality that;t can be uscddirectly for washingand bathing.The Geothermalenergy is rtlractive in dris regardns its wide r.nge rejri i,'n r(mpda,ur(Iron lhe .). err i' l+C of applicatbns allow ihe tempe.atue drop belween Foduction and rc injeclion wells to be exploitedto the full. Uses can bc land areawould releasea uscful enersyequivrlent to - 6 rnncs .as.ddcd. where the waste waler film high rentrJerarue proc rhat embodicdin our planet\ rcservesof fbssil fuelst rhe roial essescan be tbd dle Lhoserreedirg lluids at alower lcmperature. rcstricted hear resoufce in this volume is - 3.5x106TWyr. Theorelicrlly, good a high enthalpy georhernalfield coutd be cquivalentto nosL 300rines fie fbssilfuel inveniory [41]. tn usedto power a chain of applicationsftoin rherop ro ihc botton tcmrs of the nagnilude of rcsoxrce,HDR is secondonly ro of the Lindd diagram. controlled nuclear lusion in being the.on renewablecnergy This general strategy of nraxinising ttrc benefits of the oplioD fd tfie tuture and thc idea of being ablc to tap this energy geothernal loop is illustraiednt fig. 7. Specific examplesof at any hcde just by drilling deep enough is n particutarly cascadingnight bc whercrhe warcf separatedfiom flash cycte electricitygeneration ispassed io a bnr{.y cycle gencrator(figs. The crucial technical breakthroughsrequired for the eco- 4b + 4c) o. usedfbr dnrct fiearingol-habitars aru srecnnouses nomic exploitrlion ofHDR area cuLin the costsoldccp drilling (fig.4b + 6). Many other combinationswill be possiblcthrt, if and an acceptableway of creating r ificial permeablcresef crcatively iitcgrated into a ina.rian econonry. fufher enhance vonx. These reservoirs prcfefably mrde in zones of rock at the attrxctivcnessof geolhcmal projec$ on Mars. te'nperatufesof200'C or nrore-would act as hearexchangeN much rs do geothennalaquifers. Pressuiscd water would be clrcularcd thongh via injecttun and production welts and, in ''closed cn uit" designs.ihe hcrt broughtto thc suffacewoxtd -:" Ag culture be usedto gcncrateelectricity vir { binary cyclc (seefi-q. 8).

District heating Hot Cold SecondaryFluid

Reinjection Low enthalpv- fluids Pump

Industdal . Procass Geothermal High..enthalpy ! heating Tturos heat Elect city t:tt';;;;;; ',','.', Bote lnjedion | 1l I + Fluids Bore t'.','. ' Energy

Fig.7 Cascrdnrguse of geolhemslcne.ey

1l', 1116isis1l1 6. ADVANCED IIEAT MINING CONCEPTS

Hilhc.to. this papcr hasconsidered geothemat power prodnc, tion on Man analogousb its convenlio.rl pracriceon LheEalth. Howevc., cefain noij advancedconccpts fiave been studied which ponrise ways of grcatly increasingthe accessibiljryof crustalheal resoluces:lnd of nnpro\,ingiheir grnde.Econonric feasibiliiy awaitsfu1 ther research and developmenr.but ii is nor unreasoDableb expectsuch pmgressover the decadesir wilt takc for Mms explorationto evolvc into penDancntsetrlerilenl Tlrec $ch altenrxtivesafe briefl)r discussedbelow. fig.8 Exploir io. ofHotDry Rock.Ftuid ts circulaledttr rcueh, .nd 6.1 Hot Dry Rock l,eat rccotered froD, an arliflcial zone of pcrmcabletuck.

Conventionalgeorhennal power prodlctnrn is geographicay The cenrralFoblcm is rhereforeone of creatingper.ncrbitity consrrained by its reliance o. penneablerocks and a nxLur{l wherc none exists ol lircturing rocks xL deprh in a way that water supply- Heat energy however is :rvailabte anywherc opiirniscstheir heattanslera.d fluid transmissivitycharacrer bencath planet's a sufnce if its c sr is penetratedro sufticienr islics. Teclrni.tuesof doing this a.e cuffenrly being experi- deplhs.For every site groxndwater where condirionsare favour- 'nenled willr .tnd include [41]: able, there are nany more where rlrc rocks benearh may be hor. but which are also relatively impenncable and dry. An eco- (t) H tuulicprerturintion:pumpi.gwarerunderpressure nomic way ofexploili.g this "Hot Dry Rock" (HDR) rcsource to "jack" opcn natural loinis. va-sttyennances the scopeand potertial of hcat midng 141,501. For exarnplc,;t hasbeen calculared rhai an rve.age cooljng by (2) biplolive lracturins: detonating.t chenricalexptosive I'C of the iop I 0 km of crust bcneati rwo thi s of the Eanh.s .)1nDdest intensjty 1ogcnerate radial cracks. Ih. t-tilitt.fa;!.n1u1r.l Ent(' rr l'1tr5

(3) Rntid sds rt"se,ivnit,t: ignjljng a fn)tcllot or gas troirl. One pml)()scds)lulion was to deinr{tc Lhccxplosiles gcDc.rting syden. Nithin sxk doDrcsrs lhesea1t anhtdtrus rDd dllow isofiennal 'C slorqe of hcal rt N00.4 the rncltnrg poinr of sali l5ll. t4) Zrsr,dl rr"-\J//rti thc nriccrionolpnlses ofhol xnd,/of Howcvcr extraciionof hert aLthis high r gradeapFafed oDl) cold flrlds. ro bc possibleif rhe ftxrf of lhc c.r!it,\,remained Irrgely irrrcl .rnd iroduced working fluids (rn. H,. CCl, tn.l \r.r'n $re The first r\rc nrethods.ltt prcsenil) receilin: thc grcatc( \ariousl,vsug:ested) had utuesldcled access to r lalc olrnolFn .l i.." eqrr'

A nucleal explosile .rDnot be tadil) used td othef than destNcriveends as its encr?) tlease is alnrostinsttntalcoui. i This dilTiculty can bc o\ercoor howevef if this cnergy is L.'',';. |. i :'--,: . - ,. convefied to h.at ud sLored,so it .aD b€ trpped at a nloie , |-., - gndurl Lc.Rocki ltorc henr well and thus lhe idea behitld Dnclcargcothermil conceptsof the fir\L kird was io usebufied Duclcar clctonalionsto excn!rte cr\ilies .rnd deposil llrge anxruntsof lhe]:Irlalenergy at \aic deplhs.whereupon ir cotrld be erploiied in nlore or lcssconve tional \rays by intrcdtrcilg a \rorkng fluid. Sincc thc cost of a nuclear de\i.c is onl] .r shrllow lunciion of clechicrl generntionusing lhc ie ),i.ld. .H1i sidual heat from erplosio s of > I Mt wxs c{lcrlaled to be P-Faq!oacriv6 comperitjvewiih ftxsil luels. The pmblem with thi! conceptis d t thc thcmodlnrmic qualiry ofthe fcscr\of \r ould degrade rapidl) in thepresenceofgr nrdwrLcr fieloss ofheaiby mxss tmnsler lo sxrroundlng cooler sLfru causirg lhe lemper{tNe Fie. 9 R.coverl of heat fion, rhe ,ubble chinney crcllcd by r wiihin ihe high renperaturc zone to rapidly fall ro the boiling co.rained nocletu detondtilnl - cavrty 70m in radius.which wouldcolapsc forning a bbte lhe lop 10 knr of crusl. Thcy afe more common noweverr. me - chimney 300 in highi surroxndingrhis centralstru*ure, rock lower 5 km as ther bnoyancy. and hencc rate of asccnt, wonld be extensilely fracturedoul ro a disranceof- 200 m. The incfeasesas they fise dne 1()cxpansion of rhci. volume and total volume of lhe pernrcablezom would thus be - 4x101ni, cxsolur.ionof gases. their reslricred heat resource has been - from which 620 MWtt of energy, of a grade adequate for coDservarivelyesLimated to be - 7000]'Wyr, nnch ofthis being geoelec[icity.wou]d bc available.Only I 27 MWyt of rhjshear concentrntedin tbc geologicalenvironment nem rcctoric plare would comc fmm thenuclcd deviceitself, a fmction oi ttretotal bourdafies l4ll. On Mars, we mighr rcasonabtyexpcct rcces- - of 20./..Heat recoveryand powef generalbr would be ca.ded siblc magma bodies to be much rarcr, bur not neccssarity out in aclosedcncuitbyrecyclingpressudsed water ihrough the complctely absert as magmaric activiry m.Ly not be wholly rscfvoif and running it tnrorgh heatexchangc$ and a binffy extincr(lis discussedin Scction3). Any of rhc localitieswithin electricgenerator at the surface.In rhis way, the risk of sufacc the anornakNsheat tlow province deraitedin Table 2 could radioactivecontamination via the working fluid (ilself modesr potenlialy bc undeflain by ll nhlren. or parially molten, since most fission Foduct$ would be vit fied and insolubte) '1)ck as.dxring thc edrly stagesof a volcanic cyclc. nagma is would bc reduceds1ill furfier. nrtrudedinro thc shallow subsurfice(as shown in fig. 3b). Bul lbr the hysieria lbat surounds Lhcvery menlion of rhe Atfi|lst sight,itis difficulr to inaginc rhepeneralion ofmolten wo.d nxclcrr" at the presc.t rime. one is temptedto rccom rock with any eqLripmentthat coxld rcmain funciional. How, 'ncnd tufher researchinto nuclear technique! of silnutating evcr,thetechnology todothis is underdevelopmenr and lxvaat gcotherlrlalpower productionon MaIs. lf geolhennaiheat is nor 1100'C in Hawaii hasalrendy been salcly peneratedro a dcpth piisent, then nuclear explosivescan emplaceit. lf permeablc of 33 n.This i! doneby projecringvery high velocitywaterjers focks are notprescntat a given hlpcrhernal sir,e,rhen nucte.r in advanceolthe drjll encasingit within a chilledfingerofrock. exptosivescan creatcthese too. Nuclen. explosivc fraduring So long as thc bore walts arenrxintaired at below 700 .C, ir is night nrakeitteasiblc to cifcularefluids thfough layersdeeper estimaredlhat they would be sLabledo\rn ro depthsof - l0 km. than 10 km. where a ncar nrexhanstiblcHDR energyresource The curent concept of magrna exploiiation involves the w'11be avxilable anywhercon Mars. crcatjor of an open down hole heat cxchanger This wonld be donc by injectingwatcl into lhe magna whereir woutd fom a 6.3 Magma Exploitation bnlb of solidified, but hiehly pemeable rock with a lxrge sufacc area fof heat tr.rNfcl Water would flash to steam in this jr Wherc nasma chambem lie at accessibledeplhs may bc Tone and would be removed vix a pipe conccntric with the possibleto tap thcir heat contenl direcdy. as opposedto the inJectionbo.c (seefig. l0). As conveciion wl&nr $e magma coN'entunal placticcof recove.ingit rfter it hasconductedinio chamberwould he promotedb)r this process,rhc mosrefficient superposedsolid rocks f4l 551.Therc afe five distinct advan- way to operatcthe systemwoxld bc to c;rculnte$,aref ar a rare taSesof doing drjs. whcrc the wall thicknessof rhe heal excfiangertends io a conslrnt and heai is suppliedat the samcrate that the wofting (1) The cooling ofjust 1 kml of nolten ro.k frcm 1200.C lluid .cmoves ir. Attainable power ourpursper well of 25 45 'C ' to 60 releases 82,000 MWlt Although sxch a MWe have beenestimated 1551. degreeof hcat extractionwould be impracticat.magma neveftheless represenis an enonnous local coDcenrrution +water ol energy. Steam 12) Dependingon irs chemicalcomposition. rock mclts arc lound at temperaturcbetween 600 - I 100 'C. Magtnatic heat is thefefore of ve.y high gradc compared 10 convelrlionalgeothermal soufces. This would pemrit geoelecr'ii)!eneratiof r l'r! | etr.L.enr:e,.

(3) Becausemagma can transferheat by convection.very high heat erlracr;on ratesmay be possibte.

(4) Magnapossessesan additional chemicat energy potentjal e4urvalenttoaboul 1/i2 irs rhermalcnerSy porenrixt.lt can thereforcbe usedlo manufacture ', a variety offucts. :: Pemeable ',', uor instance.if rhe magm.i is dch in Fer., then this will , ' . . reduce injccted water io hydrcgen: bulb ot rock'

2FcO+H:O)Fezor+4 (.i)

Dcpendingon the down-holelcmperarwe and presiule. other reactionsare possibleby iniecting biomass with the walcr which will produce CO and CHj. //..5 Magma |i!/ (5) Some magrna chambc* can be foxnd beneath setf evidentlocations, such as active or domanr vorcanoes_ (Crypiovolcanic intrusions however woutd require greater exploraloD/ effort to idcntify and locate.) lig. l0 A downholehear exclllger cMred by irjecringwxrer innr magma.stcam ls recolertd lia a pipe colcemnc with rtp tljecln[ On the E th. many magmachambers air knownro occur in Th. utilir)' .f (i.ath.uMl r:netar.n Mo's

Because oi tlrc lower marlian gravlty. buoyancy torces are stal1investigatirg: 3- 1 9. of the planet'ssuface appearsto be of lessand lhus magma chanbers that reacb siallow depthsw ill be the yonDgcstgcological age and of volcanic or fluvial oigiD; systematicalylarger than thoseon the Eafth. Any discoveryof dresc rccks outcrop wilhin a larger prcvince compdsing lhe an accessible direct magma resource on Mrrs thcrcforc is likely €gions of Elysium. Amazoniq Areadir ald Thffsis. As these to providea long terln poientialfo. gcnckting nany megawalls fealures aI occur in just one quadrxDt of Mars- lhey nlay be of drermal. electrical and chcnical cncrgy. As with HDR genetically linted to the pfesenccof a rnanlle super-plume however, the ecoDonricsof ciploiling magma energy will below Thus. there is a potenLial for lhe discovery of dcpcndpa.tiy on a good walersupply (udess CO) sufficesas a cryptovolcanichotspots anywhere beneath this region a huge working nuid)- Thc simple lad that it is easier io transport water expanseof land, .oughly a quarterof rhe planeiaryarea. downhill, and thal any water table will be closer to the suftace Pioneering manian seltlementswill undoubiedly rcly on in basin areas,suggest ftat power production centresfor rll substantialcneryy subsidiesfrom the Ea!1hin the nranlcr ol advarced heat mining concepts will be favourcd at krw krpo boLhin1porled tuels and generationinfrastructurc. Howevet in graphicelerations. Hotspot! beneathLhc Iowlands of Arcrdia order io becomeself reliant ovcr the long-lerm,minian civdr- Planitia and the CerberusI'lains (betweenxbout I 1() 2 km salionwill have to move towardsexploiting indigenousenergy below the planetarymean rrdius) mry thus be betterplaces to flures with donestic technobgy. The greater lhe choice of start heatmining thaDanywhere on ihe heightsof Tharsis. energy sources, the morc secure sei e|s can feel about their

7. CONCLI]SIONS Geothermal eneryy may well be one of these availnble optiors.It will nol be Factical to hamessatthe very begnrning Theorelical models of global heat flow nfe not adequate as a of scltlcment,neither might it be a maiof econonic leaNre of basis to le out the present day exisrenceof gcoihcrmal a rnature planeiary civilisation but as a stcpping stone between hoispots on Mars. lndeed. geomofphobgical evideDccis pcr thc l\{,o ii could play an impoftant rclc. Sc(lements sited in hapssupportive oftheir realismas volcanicand fl vial activily geothermal areas would havc access to an abundant energy appeds to have been ongoiDg albcil al a declinhg rate-into supply that. in sone ways, is rnore reliable. efiicient and secure relatively recent geologic.tltime. thanthat gathercdfrom the sunorwind. Not only willelechical Thc economic exploil.rlion of geothermal energy on Mars is power be m{dc available,bul also a supply of hot water lor a thcrelorc a dislincl possibiliry.The conceptdepends crucially multitudc of direcl uses. thoughon nanowing down the order ofmagnitude uncetainty Somc of the fust pemanent maftian conmunities conld iniheabsolxteage of lhenosirecentmagmatism.Hypcdhcmrxl sping up aroundsuch geothernal oascs. By thelailer half of the areasbronghl aboul by activity lessthan - l0 Myr old, associ ncxl century the spatowns of Mars might cven be known lor aledwith a goundwater supply,willbe necdcdIor thc convcn ofiering the best of life o. th€ high lioniier tional lrcduction of geolhcrmal power Exploitation of areas of lesserheat flow nrxy bc possible,but will require the solt of 8. ACKNOWLEDC!]MXNTS tcchnobgy currcntl] underdevelopment forexploiting Hot Dry Dr StephenGillett is thankedfbr his hospitality in Nevnda.A Rock nceper wclls andpossibly lhe creationofa(ificial zones chancedrive past tbc SteamboatHills geothermalpower plant ofpeineability throurlr which to cjrculalean introduccdwo* outside ReDo.in i1s stark Mars like setting. was thc originnl ing fluid. inspnationfor this work.lwould liketo thank Drs. Chrislopher oI Mars will reveal its ODly a fiorough on site exploration McKay and RobertZubrin alsofor reviewsofthis paper'stirst geothermal potential. Aheady, thc.c rrc clues of where to true drtf1.

RI]FERENCES

'in B.C.Clad! ClEnistD ol theManid Surrte: Resomcs f.' ftc Minrcd Ma^- I Lewis,M.s. Matthc$s and M.L. Gueden(Flls.). RA,rzzJ ,/ Exrlu i{r orMrn," in PJ.Boson (Ed.).Ifi. cd.p/M,rs,AAS Science NedrFaih 5N.., pp. 845 885, llniveuitj' .l Anzrnn Pas. Tncson rnd Technolog!s.ncs. s7. pp.197'208, UdveltIrc., SanDiego (1984) (19931. 2. tR Mcycfdrd C.PMcKay, "TheResouresolM6 lorHund Sedencir. ' l:l. s.l,rP,lrcr,!dr.7l"er,ConFe*.llhcurilcd StatesOtriceoffechnology J6l5_42, 147160 0989). A$essndnl.cal,loenc no: 3r-600129 (1981) 3 DA Bakd rnd RM. Zubrin.'Mxs Dircdr ConbnrnrgNerTenn 14 A S. Bxrclrclof,Gealh.nnal R.soulcesC.Lnril ||eh tage, httpll Tc.lrnologicsroA.hicvc alwo Laurch MannedMas Mi$nrn,"7al.t, 43, Fev.denon.co.uvgeosci/igdhonE.lhl. 'The 5i9 525 0990). 15. C. Facc4 Shucr€ xnd Bchariour ol Ceothenn lields," in C.H. MJ Fogg, TheLoneTem H$ibbiliry of Mtrs, prcsotcdat ah. rd Anste^ l.Ed.).clrLh2rhal Lnc,"ry. pp 61 69,IJNESCO.P,ns (1973). F'PFmalc, 'Manian Volatiles: Then Degrsing Hisrory and Geochenicd 5. E l Odnn,Envitunm."t, PoN./ an.lSo. i.t, wi)ey Imerscien.e( l97l F.te.' l.dBs, 23. 179202 (1976). l. _lhelftl T.B. Jolursson,E. Kelly, AK.N Reddy.dd R.H. williams. (Lrds.). 11. MN. To*raz and A.t Hsi. llislory ud Evolurion of Mds. i.,ewarl. Er.lg). lslrd Pre$,London (1993) Iiu.L!. 34.537-s47 (1978). T.R Meyerdd C.P McKay, "IhcAinosphsr of M.6 Resourcesfor the c.FDavicsatrdR.EArvidson,MaiidnThmalHistory,corcSelEgatior, Eaploratiorand scrtlenenr of M.rs, inPJ.BononGd) Tftr Ca€r1rr atrdTeclorics- /.aas,45,339 346(t981). Mdd,AAS ScjerceandTechnrhgy Sen€s.57.!p. 209 230.Udvctlhc, 19. JW Reeder. Formationof GeodremalRe$urces d LithosFhcric snbdnction zoms," iI M.I Econ.frid€s .M Po UnrcMch (Eds.), M. Nclson. .r r/ . Usirg a CLosedEcologic{l Systenro Sludy Edrh s A liedGeoth.t,ii.s-w 117.Johnwilcy d Sons.Chichester (lt87). BiospheF."rJi,s.ic,.d, ,13, 225 236(19t3) 24. K.L. Tanala, D 1l S.ot ud R. Grcley, Clobal Straligraphy."in H.H. J.R.lrench,Rocket Prcpellrnls fron Maiian Rcsourccs.JrlS.,t2.167- Kieffdr.B.M. Jakost!, C.W Snyderand M.S Mrtlhews(Eds.). Ma,r. pp 345-382,Unire6ilr ofAiizon! r'EJs.Tucson (1992). Y lslikaea. T. Ohtita andY AnEmiya- Man Habitarion2057: Concetl 2t. M tl. Cdr, Mars:A wlbr Ri.h Planet?L2fa, 56, 187-216(1986). Designof a Mas Senlenentin theYef 2057."lal.', 43,505 512 ( 19901. 22. S\!. SqulEs stal., Icc in lhc Mdian Regolith,in1lH Kielldr,B.M. (n s-r-

421 Hydmgeologyol Mnrs, in J. Lewis. M.S. M hew\ md M.L. clerien M.H. Cdr.--FonnatiotrofMdian FloodFernrcs by Rcl.xs. olvme. fronr (*l\.).R.sautus ol NlarFarth Sl,a.c,pp.765 ?97. Universily ofArizonr contincd .luil€i " J O.,rr_ri.,'i.r., &l(|]6).2995 3007 (1979). HCH. Arndcad tud J.W Tester,H.dr firnrns. E. & FN. SponLrd, S M. Cliford. A Model for the rlydrologi. ann Clilnaic Bchaln,ur of waleron Mds, .lc..?hrr l.t.-98(E6).10973 Il016(1993). 42. C.l]rc||nettl. hrth L^\natni.s,Il.dlFLN anl Mnd. Conve.lio,5237. 25 6.Ncukum{ndD U.Wisc. M.rs:A Stlndrd Cnt$ Crrvedd Posiblc Blocf,4, OpcnUnive ny Pre\s,Milton Kcyncs ( 193| ) NewTimc S.!le," S.ra.?. 194,l3ll I l3ll7(19?6). PO UngeDach. An lntodu.tion to CeothermalEn€rgy," ln M.l. WKH.dnxnnrrrr. ChonologyolPlanctrlvolcanismbyConDnndve E.on.rrides rnd PO. frnscnxc i l.Eits.),Arpli.d c..ih{niLs. pt I a StLdierof PlurcrtuyCrdenng." in Rasahi.U.lcai:n.n th.TatF\titt lobl Wilcl'aIld S.ns. Chichener (1987). Pl,,rrr. pr 10491127, Pdganon, NewYorkO9Et). C G.?almernri, Ccorhdtul Energl,.nrTB.Johansson H Kell)-A.I.'N. 21 J.C. Lant. The Shergoil CoNontun id rhe SNC Mclconres:An Reddy,andR.E.Wilidns. (Eds.). Re,.wdrL r',rf3!, !p. 549591,Isrand oveniew, cz,.nlt. Crsn..liz. A.e,, s0.875888 (1986). 28. ABlnin, B.C.CLdk and H.Wiinte, Sud.ceChdnistry ftdMnreraologX.' E Wood, ccorh.ihrl bwer," nr CH. Armsread(Fd.), 6cal?,?'ur rH.H. KierturB.M.lakoslltr. c.w snlder dMs.M hews(Ed!.), E k'3), ru. 109121. UNEsco. Prns(19t3). ,\rarr,pp. 59'l-625. 1Jnive^itl oiArizona Prcss, Tucson (19921. PO.Ungc.rch. Electic 'Ihe PowerCcNrxini lronrceothelmL Sourccs,,-in K.L Tanaln .r d/.. Resurftcins Ilistory oi Mars: A Synthcsh of M j. Licononridcsand PO. UngeDracn(!ds.), ?.11?lic.lc.,t .r,t.r. pp. Dignired.MkingBascdccolog!, Pn.. LuMt Pltu(l. S.:iConJ,tA-665- l:7 187,loin wilcy.nd s.ns, Chichener(lgsr). (1983). 678 S.S.Einr $n, lleodrcrnnlDiin.r Heatlng-h C.H.Arnslead Gd.), 30. D H. Scot. K.L.Tanxk!. R OEeIe\ xra LE.c@rlAato1i. MLr:.tthr G.oLhlntulEntryr, Dp. 123 t34, UNESCO.pans (1973) W^ier1 F4mtutnl, Eunan Equ.turial ann P.tut R?gnn\ .J Mds, Mats 18 B Lnrdil, lndu$nal and Other Applications of ce.rhemral EnerEy, in I-1802A. B andC,fr.S. Ccological Surver (1986-87). CH. Amrste.d(rid.). O.,rr!,,r/ E !c,Al',pp. ll5-148, UNESCO,Pds LB. Plcs.ia, "Reent lLood Lavasin dE Elysimr Regionol Mds.- t rryr, I S. Cndmuds*)nrnd l.W Lund. Dir.i Usesof Eafthlled, ir M.J. J.B.Plesci!, -r2. A[\sscssncntotv()laLiteReleasef|omRccenrVrtcanisrnin Econonidesud Po. U.genuch(Edi.), A4,lr.l a;r,/is,,r'r1, l)p 189 Elysnun, Mr\." /cdm, 104, 20,32 ( | 993) 219,J(,hn \!ileJl ald sons.chiche{er (t98t). -in PI.Moulirh Marl,I-Wllson iLWHcadIlI. Explosirevol.anisnion 50 A.S.B chcl.r "HorDry ltock Ljxploitaiio., M J.EconoDidcs and pO HecatesTholus. MaFi InvestigationofEnption Coidnnrs," J. c?r//.r. Unsenach (Eds.). t?ftnd Ge.tr.,,i.r. pp. 221 23,1.lohn lviley and r?.r..37iB12), 9890 9904 lt9rJ2).'lgdmbincs D.ll. Sconand K L Tan.kr rI An!7onisPladtia 01 Mars,.. 51. H.G.Comeil and A.D. Sufi1e.Usc of Pn)n Nucle.rDeviccs tur the t. (1932) A..rt/])s.Res.. A7l.n2) l lt9 l l90 It{ore4' of lrrcrnd Energy,' in Ptu...linls 4 tt2 Saoa,l t'lowhaE PJ.MouginiFM.rk, RcccntwarerRele!senriheTha$isRegionotM.r,.. S)rVosiun, Pan lII, UCRL56?7, pp. 51-57. US Aronrir Energy l.?,2r.84,3623t3 (1990). CB.FurnuandPE Don$- Clo b.l Scasoi al van dli(,n oivarervxpolr on 52. R D. ForsytherndIR 7imbelnm, ACascfnrAncienrEvapoLirellasiN Ma and the hrplicarions br Pendrost,' J. C,prrr Rer, 84, 288l onMtus. r 6.0/r:r. Rsr..100(Fi3), 5553 5563 (1995). 53. GC Kennedy,AProposal loi x Nlclear Power -Recent 'S.r,.r.?. PrcemD, iI Pzl.r./i,Sr B.L Luchitu. MaficVolcmisn on M!r, 235,565 56? .l Lht Thral PltNthare StnpasiLr, TID 7695,lp. 305 308. US Atoniic { 98?). hne4y co.misstun (l 964) B.K LLchilta, YoungVol.rnic Deposits hthe VxllcsMdnen!- Mas?' I B BumhamlndD.H.Stc*an.TheEcononicsolPloNshdeC*orhemal /.dfas,36. 4?6 50!l (1t90). PoN.r, in E ,gtr.rjn8 nt 1va.Laftyr,rir4 CONF700101 (Vol.2), 39. vR.Baker.r/. nn.ienrOcelrs,Ice Snetsand rheHy&olosicalCyctc pp.1376-1383.US rmicEnergtCondssion(1970). _fhc on Mr!" Narru, 352,539 594{1991) 55 TY Chu.r,1, MagnaEn€rgy Research Progmrn.- ?idrd./r,s,l theceothetflul R.soutr.s c.tttil, 1:lO). 56757? 0990).

122