Energetic Neutral Atom Imaging of the Heliospheric Boundary Region

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 106, NO. A8, PAGES 15,767-15,781, AUGUST 1, 2001

Energeticneutral atom imaging of the heliospheric boundary region

Mike Gruntman,• Edmond C. Roelof,2 Donald G. Mitchell,2 Hans J. Fahr,3 HerbertO. Funsten,4 and David J. McComass

Abstract.Energetic neutral atom (ENA) imagingis a powerfultechnique, which can remotely probethe properties of distanthot plasmas. Hot plasmas are abundant at theheliospheric boundary,the region where the expanding solar wind meets the surrounding local interstellar cloud.Here we presenta newconcept for imaging this boundary in ENA fluxes.Heliospheric ENAsare born from charge exchange between energetic protons and the background interstellaratomic hydrogen gas. The technique is idealfor studyingthe asymmetric three- dimensionalheliospheric interface region remotely, from 1 AU. We showthat ENA imagingin the0.2-6 keV energyrange will establishthe nature of thetelTnination shock and properties of hotproton populations in theheliosheath. We alsoexamine how the evolution of pickupproton populationsat andbeyond the shock can be explored.Global heliosphere ENA imageswill distinguishamong the competingmodels of the interactionbetween the Sunand the local interstellarmedium, and they will revealthe physics of importantprocesses in theinterface region.Heliospheric ENA fluxesare exceptionally weak, which makes imaging implementation difficult.Nonetheless, we showhow single-pixel ENA sensorscan image the heliosphere from a spicmingspacecraft on a typicalmission near 1 AU. Therequired instrumentation is briefly discussed.

1. Heliosphere equal the externalLISM pressureof the interstellarwind and magnetic field. The solar wind expansiontransitions to a The interactionof the expandingsolar wind plasmaand local subsonicflow at thetermination shock. There the kinetic energy interstellarmedium (LISM) createsthe heliosphere[Davis, of the supersonicflow is largelyconverted into thermalenergy 1955; Parker, 1963; Dessler, 1967; Axford, 1972]. The in the subsonicplasma beyond the shock. This subsonic heliosphereis a complexphenomenon where the solarwind and postshocksolar wind plasma flows around the termination interstellar plasmas, interstellar gas, magnetic fields, and shock and down the heliospherictail. The solar plasma energeticparticles all play importantroles. eventuallymixes with the interstellargalactic plasma in the tail The Sun's interactionwith the nearbyinterstellar medium is at distances of 1000-2000 AU. of considerableastrophysical interest. Our Sun is typicalof its Interstellarplasma does not penetrate through the heliopause, classof stars.Thus the hellosphereoffers a uniqueopportunity the discontinuoustransition boundary separating the solarand to study in detail the only accessible example of a galacticplasmas, but flows aroundthe hellopause.The region commonplaceand fundamentalastrophysical phenomenon, the betweenthe terminationshock and the hellopauseis calledthe formation of an astrosphere.Wood and Linsky [1998] heliosphericsheath (heliosheath). Other hellosphericmodels demonstratedthe possibility of remotely observing the make somewhatdifferent assumptions,including a subsonic astrospheresof nearby stars,which opensthe possibilitiesof interstellarwind, a weak terminationshock, an asymmetric comparativeastrosphere studies. solarwind, and may includean interstellarmagnetic field and A possible two-shock Sun-LISM interaction scenario cosmicrays [e.g., Fahr and Fichtner, 1991; Suess,1993; Linde [Baranov,1990; Baranovand Malama, 1993, 1995] illustrates et al., 1998; Zank, 1999]. In all the models,the hot shocked the main featuresof the hellosphere(Figure 1). The interstellar solar wind plasma fills the heliosheathand flows down the wind approachesthe hellosphere with a supersonicvelocity and hellospherictail. formsthe bow shock.The dynamicpressure of the expanding, Three major factors make the heliosphere essentially highly supersonicsolar wind decreaseswith the heliocenthc asymmetric.First, the partially ionized interstellargas of the distance.At a certaindistance from the Sunthis pressure would [.ISM moveswith a 26 km s-• speed(Figure 1) withrespect to the sun (interstellarwind). Second,the interstellarmagnetic field couldbe pointedin any direction.Third, the solarwind is anisotropic with the properties over the Sun's poles of lDepartmentSouthernCalifornia, ofAerospace Los Angeles, andMechanicalCalifornia. Engineering, Universitysignificantly differentfrom those closeto the ecliptic plane. 2AppliedPhysics Laboratory, Johns Hopkins University, Laurel, The hellospherehas been extensivelystudied both Maryland. theoreticallyand by variousexperimental techniques since the Universitiit3InstitutBonn, far AstrophysikBonn, Germany. und Extraterrestrische Forschung,early 1960s. Experimental dataon the region between the 4LosAlamos National Laboratory, Los Alamos, New Mexico. supersonicsolar wind and the surroundingLISM, the 5SouthwestResearch Institute, San Antonio, Texas. hellosphericinterface region, are scarce,mostly indirect, and often ambiguous.Although significantprogress was achieved Copyright2001 by theAmerican Geophysical Union in understandingthe Sun-LISM interaction,many important Papernumber 2000JA000328. questionsare still unanswered,for example:Is the solar wind 0148-0227/01/2000JA000328509.00 expansionterminated by a shock?What is the nature of the

15,767 15,768 GRUNTMAN ET AL.: ENERGETIC NEUTRAL ATOM IMAGING OF THE HELIOSPHERE

BS Becausethe terminationshock and the interfaceregion beyond are so huge (hundreds of astronomicalunits), only remote techniquescan provide a global view of the time-varyingthree- dimensionalhellosphere on a continuousbasis. We argue that the technique for obtaining our first informationon the globalstructure of the heliosphereis already ISG at hand. This technique, energetic neutral atom (ENA) imaging, is sensitiveto the details of the structureof the s hellosphericinterface, to the nature of the terminationshock, M and to the evolutionof the pickup proton propertiesat and beyondthe terminationshock. The scientific promise of ENA imaging has been increasinglyargued for since the early 1980s and has led to extensivedevelopment of the conceptand instrumentation.The conceptwas validated by imaging the magnetosphericion Figure 1. Two-shockmodel of the interactionof the solarwind populationsusing ad hoc techniqueswith existingnon-ENA with the local interstellarmedium. Angle © is measuredfrom instruments[Roelof, 1987; Henderson et al., 1997] and the upwind direction. The hatched region between the dedicatedENA instruments[Orsini et al., 1992; Mitchell et al., terminationshock and the heliopausecontains the postshock 1993, 2000; Roelofet al., 1999;Brandt et al., 1999;Krimigis et solar wind plasma and pickup protons.The energeticneutral atoms (ENAs) producedin this region are the focus of this al., 2001; Pollock et al., 2000; Moore et al., 2000]. Williamset work. LISM, local interstellar medium; TS, termination shock; al. [1992] reviewed the conceptof magnetosphericimaging, HP, heliopause; BS, bow shock; CR, cosmicrays; ISP(G), and Gruntman [1997] reviewed ENA instrumentationand interstellarplasma (gas); B, magneticfield. presenteda historyof ENA experimentalstudy. Imaging of the hellospherein ENA fluxes has been advocatedfor sometime [Gruntmanet al., 1990; Gruntman, 1992, 1997; Hsieh and Gruntman,1993] as a way of exploringremotely properties of termination shock? What is the distance to the shock and its the hellosphericsheath plasma. The first dedicatedspace shape?Is the natureof the shockthe samein all directionsfrom experimentto detectthe heliosphericENAs wasprepared in the the upwind (with respect to the interstellarwind) to the mid-1980s[Gruntman et al., 1990] but has neverbeen flown. downwind?What happensto the solarwind pickupions at and ComplementaryEUV mapping of the hellopauseis being beyond the shock? We also know very little about the currently evaluated [Gruntman and Fahr, 1998, 2000; heliopauseand the bow shock. Gruntman,2001 a, 2001b]. Inferring the properties of the heliosphericinterface is In this articlewe presentthe conceptof heliosphereimaging complicatedby the complexity of the Sun-LISM interaction. using naturally occurring ENA fluxes. ENA imaging will The plasmaand neutral gas flows aremultifluid with oftennon- distinguish among the competing Sun-LISM interaction Maxwellian ion velocity distributions.Plasmas and neutral models,radically constrain the models,and explorethe physics atompopulations are not in thermodynamicequilibrium, many of the criticaland often superimposedprocesses in the interface processesare nonstationary,energetic ions substantiallymodify region.Heliospheric ENA fluxesare exceptionallyweak, which the shocks,and galacticand anomalouscosmic rays contribute makes imaging implementationnontrivial. Thereforewe also importantpressure gradients. The mean free path of neutral describehow the hellospherecan be imagedon a realisticspace atoms is comparableto the size of the heliosphere,which missionbased on the availableinstrumentation technology. strongly limits applicability of gasdynamical methods. Consequently,a self-consistentmodel of the stationary 2. Heliospheric Interface heliospherehas yet to be developed. The lack of the direct experimentaldata severelylimits our The propertiesof the LISM determineimportant features of abilityto understandthe detailsof the hellosphericinteractions. the Sun-LISM interaction.The physicsof the LISM is far from Furtherprogress in studyingthe Sun-LISM interactioncritically completelyunderstood [Cox and Reynolds,1987; Frisch, 1995; dependsupon the validation of the fundamentalassumptions Breitschwerdt,1996]. The LISM is not in thermodynamic underlyingthe conceptof the heliosphere.Direct and "clean" equilibriumand may not be in a stationarystate, because of the experimentaldata on the nature of the terminationshock, its Sun's proximity to the edge of the Local InterstellarCloud distanceand shape,and the evolution of the pickup proton [Linskyet al., 2000]. The LISM velocityvector with respectto populationwill substantiallyadvance our understandingof the theSun (26 km s-';ecliptic longitude 252 ø andlatitude +7 ø) and heliosphereand constrain the competingtheoretical models. the LISM temperature(-7000 K) havebeen reliably established Within the next several years, Voyager 1 may reach the [Lallement et al., 1990; Witte et al., 1996]. However, the terminationshock. If so, the spacecraft,now nearing80 AU, inferrednumber density and the degreeto which the LISM is will establish the distance from the Sun to the termination shock ionized are poorly known; we also know very little aboutthe in one point directionfor a particularphase of the solarcycle. directionand magnitude of the interstellarmagnetic field. Voyager 1 is headedin approximatelythe upwind direction, Interstellarplasma and neutralgas are efficientlycoupled in and the termination shock is estimated to be somewhere the heliosphericinterface through charge exchange,which between 80 and 100 AU [e.g., Lee, 1996]. The planned modifies the properties of the inflowing interstellaratoms InterstellarProbe spacecraftwill crossthe terminationshock, [Wallis, 1975; Ripken and Fahr, 1983; Baranov and Malama, explore the heliosphericinterface, and samplethe LISM in 1993, 1995; Lallement et al., 1993; Fahr et al., 1995; Fahr, approximatelythe samedirection some time afterthe year2020. 1996; Izmodenov et al., 1997]. The most prominent GRUNTMAN ET AL.' ENERGETIC NEUTRAL ATOM IMAGING OF THE HELIOSPHERE 15,769

Fahr and Ziemkiewicz,1988; Bogdanet al., 1991; Yoonet al., i Hydrogen"wall" 1991; Chalov et al., 1995; Chalov and Fahr, 1996; Fahr et al., -3 Wind 2000]. Pickupions provide seed particles [Fisk et al., 1974] for FIH,cm. '-'"-• Interstellarion acceleration [Pesses et al., 1981] to cosmicray energiesat the terminationshock, forming the so-calledanomalous cosmic 0.2 .__ rays. The shock itself is likely modified by pickup ion accelerationin the precursorand subshockregions [dokipii and Giacalone, 1996; Fisk, 1996; Zank et al., 1996]. The pickup protonswould not only substantiallyweaken the shockbut also modify the energy spectraof anomalouscosmic rays in the shock'svicinity [Chalov et al., 1995, 1997; Chalov and Fahr, O.1 1996; LeRoux and Fichtner, 1997]. The lack of direct experimentaldata limits our ability to validatethe conceptsof t••x:::•'.• [ lnterstellmparticle acceleration at, andthe nature of, thetermination shock. i 0.14cnf 3 iii•!•..-[ density 3. Termination Shock We do not know what exactly happensin the region of the Heliotail• ionizationcavity il heliosphericinterface. However, we do know that the solar depletio• <4AU 1 1000 wind is supersonicat least to the present-daydistances of AU Voyager spacecraftnearing -•80 AU, and we do know that the solar wind supersonicexpansion cannot continueindefinitely Figure 2. Distributionof interstellaratomic hydrogen.The becauseof the finite LISM pressure.If the conceptof the solar density enhancement("hydrogen wall") is the effect of the wind terminationby a shock is correct,then there is a region neutral-plasmacharge-exchange coupling. The numberdensity containinghot (shocked)decelerated (subsonic) plasma beyond precipitouslydrops because of the ionization in the Sun's the terminationshock. The propertiesof the postshockplasma vicinity (solar cavity). Calculationsof Baranov and Malama would stronglydepend on the natureof the shock,i.e., whether [1993, 1995]. the shock is predominantly gasdynamicaland strong [e.g., Baranov, 1990] or moderatedby energeticparticles and weak manifestationof this plasma-gascoupling, depicted in Figure2, [Zank et al., 1993; Chalov and Fahr, 1994, 1996; Lee, 1996; is an upwind enhancement(hydrogen wall) and downwind LeRoux and Fichtner, 1997; Isenberg, 1997b; Fahr and depletionof the atomichydrogen density. The measurementsof Rucinski, 1999; Fahr and Lay, 2000]. We also do not know the hydrogenglow in the 121.6-nmresonance line confirmed how the solar wind energyis redistributedbetween the proton the predicted density enhancementand slowing of the andelectron components of the postshockplasma. interstellarwind [Lallementet al., 1993].The complexityof the Pickupprotons constitute a significantfraction, up to 20%, time-varying asymmetricSun-LISM interaction essentially of the solarwind at the terminationshock. The pressureof the limits the utility of theseglow measurementsfor revealingthe pickup ions could significantly weaken or moderate the details of the interaction. terminationshock. The pickup ions moderatethe shock,while The plasma-gascoupling also determinesthe extensionof the type I and II Fermi processesare believed to accelerate theheliospheric tail. The initiallyhot (>--106 K) shockedsolar pickupions, producingeven more energeticions, which would wind plasmaenters an extendedplasma tail. ParkerS•[1963] weaken the shock further. Though theoretical modeling original description of this plasma flow was recently providessome insight into proton injectionand accelerationat generalizedby inclusionof charge-exchangereactions with the terminationshock, all the experimentaldata are indirect. interstellarhydrogen atoms enteringthe tail from the outer The nature of the shock may also dependon the ecliptic flanks[e.g., Osterbart and Fahr, 1992;Baranov and Malama, latitude.The typical solar wind speedover the Sun's poles is 1993; Czechowskiet al., 1995; Jaeger and Fahr, 1998]. No significantlyhigher than the speedclose to the eclipticplane, experimentaldata are yet availableto test our conceptsof the 750 versus450 km s-• [Neugebauer,1999; McComas et al., heliospherictail. 2000a]. Hence an average solar wind proton carriesalmost 3 Ionization of hellosphericneutral atoms producespickup timesmore energyin the polar directionsthan near the ecliptic. ions in the solar wind [e.g., Ifolzer and Axford, 1970; Fahr, In addition,the ecliptic latitude where the transitionfrom the 1973; Vasyliunasand $iscoe, 1976; Moebius et al., 1985; high-speedto slow-speedflow occurssignificantly depends on Gloeckleret al., 1993]. The solarwind plasmaapproaching the the solar cycle phase [Bertaux et al., 1996, 1999; McComas et terminationshock is in an essentiallynonequilibrium state. The al., 2000a, 2000b]. plasmaconsists of three major distinctcomponents, original Solar wind momentumflux is •50% higher over the sun's solarwind protons,solar wind electrons,and pickup protons. poles than near the ecliptic plane. Thereforethe termination All threecomponents have the samebulk velocitybut different shockwould likely be a factor of •1.2 farther away over the effective temperatures.In addition,the pickup protonsthat Sun'spoles compared to the eclipticat the sameangle from the constitute10-20% of the solarwind at the shockare essentially upwinddirection. The solarwind pickupproton fraction at the non-Maxwellian. shockis roughlyproportional to the distanceto the termination Evolutionof the pickupion velocitydistribution function, as shock times the charge-exchangecross section.Because the the solarwind expandsto the terminationshock, and the nature cross section at 750 km/sec is a factor of-•l.2 less than that at of the termination shock are complex and not completely 450 km/sec,the pickupproton fraction would be approximately understoodphenomena [e.g., Isenberg,1986, 1997a, 1997b; the same.However, the pickupenergies would be significantly, 15,770 GRUNTMAN ET AL.: ENERGETIC NEUTRAL ATOM IMAGING OF THE HELlOSPHERE a factor of---3, different. In addition,the solarwind magnetic ENAs provide an often-ignoredmechanism of transportof field structuredramatically varies with ecliptic latitude.All energyand momentumin the heliosphere.The ENAs born in thesevariations may translateinto fundamentaldifferences of the supersonicsolar wind [Fahr, 1968b;Gruntman, 1994] and thetermination shock at differentlatitudes and longitudes. heliosheathwould penetrateand "contaminate"the pristine Gasdynamicalmodels of the interactionfail to properly LISM at distancesup to 300-400 AU [Gruntman, 1982]. describenon-Maxwellian plasma ion velocity distributions,in Therefore,even a supersonicinterstellar wind would "learn" particularthat of the pickupions. However,such models are about the heliospherebefore reachingthe bow shock. The useful for describingthe macroscopicproperties of plasma interstellarwind flow would thus be heated, slow down, and flows becausethey are fundamentallybased on conservation deflect[Gruntman, 1982] as shownin Figure 1. laws. Gasdynamicaldescriptions allow a convenientinclusion We concentratein thiswork on hydrogenENA fluxeswithin of the magneticfield effects[e.g., Linde et al., 1998;Ratkiewicz the energy range 0.2-6 keV. The dominantsource of such et al., 1998]and solar wind anisotropy[e.g., Linde et al., 1998; ENAs is the postshocksolar wind plasmaand pickup protons in Zank, 1999] and self-consistentinclusion of the plasma-neutral the heliosheath(hatched area in Figure 1). This is the most gascoupling with thekinetic description of neutralcomponents promising energy range for study of the three-dimensional [Baranov and Malama, 1993, 1995]. In the absenceof structureof the heliosphere,nature of the terminationshock, completeself-consistent interaction models, one can describe and evolutionof the pickup protonproperties. We will show the pickupproton populations as beingconvected by the flow thatthe predictedENA fluxesare high in intensityand strongly field obtainedgasdynamically. depend on the details of the physical processesin the heliosphericinterface. 4. HeliosphericEnergetic Neutral Atoms At higher energies,55-80 keV hellosphericENAs were likely detectedby the neutral channel of the High-Energy ENAs are born in the chargeexchange between energetic SuprathermalTime-of-Flight (HSTOF) sensorof the Charge, plasma ions and backgroundneutral gas. When charge Element,and IsotopeAnalysis System (CELIAS) on Solarand exchangeoccurs, the resultingENA instantaneouslybecomes HeliosphericObservatory (SOHO) [Hilchenbachet al., 1998]. independentfrom the surroundingplasma and the influencesof The fluxes of such higher-energy ENAs are orders of the magneticfield. The neutral particle moves in a ballistic magnitudesmaller than thosewith E<6 keV. However,ENAs trajectoryaway from the point of chargeexchange governed with E>20 keV could open a way [Hsiehet al., 1992;Roelof, only by the initial velocity and gravitationalforces. With the 1992] to explorethe importantprocesses of ion accelerationat exceptionof very low energies(< 50 eV/nucleon),gravitation shocksto cosmicray energies.Shocks are known to provide can be disregarded,and one can assumethat ENAs preserve efficient mechanismsfor ion accelerationin spaceplasmas thoseenergetic ion velocitiesimmediately before the charge- [Czechowskiet al., 1995; Gloeckleret al., 1994, 1995;Keppler exchangecollisions. As ENAs travelthrough space, they may et al., 1995; Chalov et al., 1997; Hilchenbach et al., 1998; be lost by ionizationin a secondcharge exchange, electron Chalov and Fahr, 2000]. The high-energyENAs are beyond collisions,and photoionization.ENA lossis usuallysmall and the scopeof thiswork. mostly occursclose (<10 AU) to the Sun; elastic collisions (scattering)of ENAs on the solarwind plasmaare negligible beyond1 AU. The ENA ability to fly straightacross magnetic 5. ENA Production and Loss field lines and their small and reliably quantifiableloss allow probingof remote,inaccessible regions in space.Additionally, The postshocksolar wind plasmaconsists mostly of protons ENA imagingprovides the only way to imageplasma protons. (• 95%) and alphaparticles (•5%), while backgroundneutral It wasunderstood in the early 1960sthat the processesat the interstellargas consistsof hydrogen (•90%) and helium solarwind terminationregion would produce ENAs (seereview (•10%). The hellosphericENAs are predominantlyhydrogen by Gruntman[ 1997]).The emergingconcept of theheliosphere atoms.The cross-sectionenergy dependences for protoncharge [Davis,1955; Parker, 1963] was extendedin 1963by ,4xfordet exchangeon hydrogenand helium arc shownin Figure 3. al. [1963], who predicted the formation of ENAs in the Charge exchangeon hydrogenwould clearly dominatethe heliosphericinterface, and by Pattersonet al. [1963], who productionof the hcliosphericENAs with energiesbelow 100 suggestedthat about half of the solar wind protonswould kcV. reenterthe solar cavity (with three fourth of the initial solar The details of the calculationsof the hellosphericENA wind velocity)in the form of hydrogenENAs. This latteridea fluxes can bc found in the work of Gruntman [1992, 1997]. was put forward to explain the then-recentlydiscovered Briefly,the directional differential ENA flux J•A,i ( cm-2 S'• sr-• [Morton and Purcell, 1962; Pattersonet al., 1963] atomic erg-•) of a givenspecies i from a givendirection • is a lineof hydrogenin interplanetaryspace. sightintegral Fahr[ 1968a]explained how interstellarneutral atoms could penetratethe hellospherefollowing the Kepleriantrajectories and reach the Sun's vicinity. Measurementsof interplanetary jEN^.,(•,E)=lj,(•,E)•[o',k(E)nk(•)]P(•,E)ds, (1) glow in the hydrogen and helium resonancelines firmly establishedthat interplanetaryspace is filled by interstellargas directly enteringthe solar system[Fahr, 1974; Holzer, 1977; wherej,(],E) is the localdirectional differential flux of Meier, 1977; Thomas,1978; Bertaux,1984]. Gruntman[1992] parentions, nk(]) is the localnumber density of neutral showedlater that the ENA flux from the hellosphericsheath is speciesk of the backgroundgas, and •k(E) is the energy- highlyanisotropic and significantlyweaker than was suggested dependentcharge-exchange cross section between ions of by Pattersonet al. [ 1963]. speciesi and neutralsof speciesk. The survivalprobability, P, GRUNTMAN ET AL.: ENERGETIC NEUTRAL ATOM IMAGING OF THE HELIOSPHERE 15,771

1 .E+02 (negativeVEN^), which couldnot be measured(Figure 5b). The transitionin the terminationshock would slowthe plasmadown 04 1.E+01 and heat the protonpopulation. Many protonswould now have E the velocitiespointed toward the Sun, and someENAs would •-. 1.E+00 be producedwithin the instrumentenergy band (Figures 5c-5e). The measuredENA flux would thus stronglydepend on the • 1.E-01 type of the shock.

o 1 .E-02 If the shock is strong (gasdynamical),then the solar wind ,_ plasmawould significantlyslow down and be stronglyheated e 1 .E-03 (Figure5c). If the shockis weak (moderatedby energeticions),

o then the velocitychange and heatingwould be small (Figure o 1 .E-04 5e). The expectedENA flux is much higher for the strong shock.A typical instrumentwould have severalenergy bands 1 .E-05 allowing measurementof the distributionof.the ENAs with 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 positivevelocities. H+ energy, eV The energyof protonsin the supersonicsolar wind is 1-3 keV, while the energy of electrons is a few eV. The Figure 3. Cross-sectionenergy dependence for protoncharge gasdynamicalmodels often treat the plasmaas a single fluid exchangeon hydrogenand helium atoms. From Gruntman assumingthe same proton and electrontemperatures. In the [1997]. caseof a strongshock, a large fractionof the supersonicsolar wind kineticenergy is transformedinto the thermalenergy of of ENAs allowingfor particleextinction on theirway fromtheir the plasma.If mostof this energyremains in the plasmaproton pointof birthto the observationpoint is component,then the protontemperature would be significantly higher than the electron temperature.This uneven energy distributionwould be pronouncedin ENA energydistributions (2) = expE- ] , (Figures5c and 5d). The bulk of the pickupprotons in the supersonicsolar wind where,8 (t) is the ENA total lossrate from chargeexchange, upstreamof the terminationshock does not producedetectable electronimpacts, and photoionization. ENAs (Figure 5f). The strengthof the terminationshock would The ionizationrate is inverselyproportional to the squareof determinethe energydistribution of the ENAs producedby the the distancefrom the Sun for R>0.2 AU. The typical ratesat 1 pickupprotons in the downstreamregion (Figures5g and 5h). AU are [3E = (4-8)x10'7 s-• [e.g.,Fahr, 1974;Holzer, 1977; We do not know what exactlyhappens with theproton velocity Gruntman, 1994; Rucinski et al., 1996; McComas et al., 1999; distribution in the shock transition, whether it remains Bertaux et al., 1999]. We assumethat the ionization rate is unchanged,being compressed,thermalized (Maxwellized), or constantin time and independentof heliographiclatitude and evolvesin some other way. The ENA energy distributions longitude.Figure 4 showsthe probabilitiesof reaching1 AU would reveal the pickup protonproperties at and beyondthe forENAs moving radially toward the Sun and [3 E = 6x10 -7 s -• as termination shock. a functionof the ENA initial (at infinity) energy. The 0.2-6 The sensitivityof ENA distributionsto proton velocity keV ENAs are of primary interestfor the presentwork. The distributionsand bulk velocitiesmakes ENAs a powerfultool to survivalprobability is -• 60% for the ENAs at the low-energy remotelystudy the propertiesof hot plasmas.Depending on the limit, and it graduallyincreases to 90% at 6 keV. directionof the shockedplasma flow in the heliosheath,the flux The differentialflux (per unit area and sold angle in unit time) of ENAs with velocity V comingradially toward the Sun to an observerat a pointR 0 is givenby 1.0

j•^ (v)= I N•(R)N•(R)f(V,R)ø'(V)P(V,Ro)VdR, (3) Ro 0.8

whereNp and N• are the local protonand neutralnumber ø06 densities,o is the velocity-dependentcharge-exchange cross f• ß section,jc(V,R) is the local protonvelocity distribution function in the observerreference frame, and P(V, Ro) is theprobability to 0.4 reachthe observationpoint withoutloss.

o 6. Remote Study of Hot Plasmas r-, 0.2

Figure5 illustratesthe conceptof remoteprobing of proton ß velocitydistributions and its sensitivityto plasmatemperatures. 0.0 I I . I I III I . I I I . II I I I I I I I I Let an instrumentmeasuring ENAs in a givenvelocity (energy) 10 100 1000 10000 band be pointedin the antisolardirection (Figure 5a). The H ENA energy, eV charge exchangein the supersonicsolar wind flow would produce ENAs with the velocities in the antisolar direction Figure 4. Survivalprobability for H ENAs to reach1 AU. 15,772 GRUNTMAN ET AL.' ENERGETIC NEUTRAL ATOM IMAG1NG OF THE HELlOSPHERE

Tem•ination ENA of ENAs at a given velocity (energy) would probe different partsof the protonvelocity distribution function and may vary shock instrumentSun by ordersof magnitude.ENA spectraare extremelysensitive to the flow patternin the heliosheath.Global ENA imagingthus gives us spectralinformation on the natureof the evolutionof the protonpopulations at the terminationshock interwoven with SolarWind... f(V) t b the globalflow patternof the plasmain the heliosheath. Upstream•. [_ Velocity The solarwind reachesthe terminationshock in ---1year. A [ I (energy) hydrogenENA with energy •-1200 eV would crossback the distanceof---100 AU from the hellosphericsheath and reach an "• '•" / band observerin the inner solarsystem also in 1 year. It wouldtake 2 yearsfor a 300-eV ENA to reachthe observer.Therefore the heliosphericimages would measurethe heliosheathplasma Heliosheathplasma a• propertiesaveraged over a few years. Thereforehellosphere Strongshock ...... '":."•::•:.•:,•I t2 imagingis capableof probingthe 11-yearsolar cycle variations Singlefluid...... •:• .... •...... of the interfaceregion.

ENA 7. Model Heliosheathplasma We demonstratethe main features and potential of Strongshock { d hellosphereimaging in the ENA fluxes using a two-shock gasdynamicalmodel [Baranov and Malama, 1993, 1995; Single-protonfluid• ,• Baranovet al., 1998]. The two-shock(Baranov) model assumes thatboth the interstellarplasma flow andthe solarwind plasma VENA flow are supersonic.The plasmasare describedas singlefluids, while the neutralgas is describedkinetically. The plasma-gas Heliosheathplasma charge-exchangecoupling is treatedself-cOnsistently. Cosmic Weak shock rays, energeticparticles, and magneticfield are disregarded. Singlefluid The hellosphericaxisymmetric plasma-gas flow fields(velocity, temperature,and number density) were calculated for thiswork VENA by V. Baranov and his group at the RussianAcademy of Sciences,Moscow. Pickupprotons The following LISM parameters(at infinity) are assumed: upstream velocity,25.0 km s-•;temperature 5672 K; electron(proton) numberdensity n• = 0.07 cm-3;neutral hydrogen number densityn. = 0.14cm 3. The solar wind is sphericallysymmetric VENA withthe speed 450 km s-• andnumber density 7.0 cm-3 at 1 AU. The calculatedstrong termination shock, heliopause, and bow shock are shown in Figure 6. The calculateddistribution of Pickupprotons neutralhydrogen was shownin Figure 2. By measuringthe Strongshock directionaldependence of ENA fluxes, one can constructa composite all-sky image of the hellosphere from inside [Gruntman,1997]. VENA Two major possibilitiesof the terminationof the solarwind expansionare a strongshock and a weak shock.In the caseof a strongshock, the two-shockgasdynamical model as described aboveis applied.The ENA fluxes are calculatedusing (3) for Picktipprotons given plasma local number densities, velocities, and Weakshock • h temperaturesalong the lines of sight. In the case of a weak shock,the bulk of the original solarwind plasmawould not be sufficientlyheated to produceENA fluxes with the energies >0.2 keV. Themost prominent source of thehigher-energy ENAs With the energiesup to a few keV is the pickup protonsin the Figure 5. Remote probing of the ion velocity distribution postshockregion. Therefore ENA globalimages would directly function f(V). (a) Observation geometry. (b) Supersonic reveal the evolutionof the pickup protonpopulation at and (upstream)solar wind. (c) Heliosheath(shocked) plasma and beyond the termination shock. The details of this evolution strongshock (single fluid). (d) Heliosheathplasma and strong shockwith all the energyconfined in the protoncomponent. (e) couldbe differentover the Sun's polesand near the ecliptic. Heliosheathplasma and weak shock (single fluid). (f) Solar We do not know what exactlyhappens to pickupprotons at wind pickup protonsupstream of the terminationshock. (g) the shockcrossing or later, when they are being evacuatedto Pickupprotons in the heliosheath(strong shock). (h) Pickup the heliospherictail. We considerhere two deliberatelysimple protonsin theheliosheath (weak shock). and qualitativelydifferent possibilities. First, we assumethat GRUNTMAN ET AL.: ENERGETIC NEUTRAL ATOM IMAGING OF THE HELIOSPHERE 15,773

1400 velocity distributionproduce numerous energetic neutrals with the energiesup to severalkeV (not seenwith the color coding used). The typical energy distributionsfor four different 1200 directionsare shown in Figure 7. Measurementof the ENA fluxes in a few bands in the 0.2-6 keV energyrange would

1000 reliably establishthe curve slopes.This range is also most importantfor probingthe pickup protonsin the heliosheath.In addition,ENAs in this energyrange are readily detectableby 8OO the existing techniques. Therefore we concentrate on heliosphereENA imagingin the 0.2-6 keV energyrange. A weak termination shock would show very different 6OO signaturesin ENA fluxes. The bulk of the originalsolar wind plasmawould only be slightly heatedby the transitionacross 4OO the shock.Such heatingis insufficientto producethe desired numbersof ENAs with the Sun-pointedvelocities. However, the pickupprotons would providean excellentmeans to probe 200 the interface region structure.Solar wind turbulencemay increasethe "diameter"of the pickupproton distribution sphere in the velocity spaceand extendthe wings of the distribution. B i ! ,, ! • i i i ß B I 400 200 0 -200 -400 -600 -800 Recentestimates [Fahr and Lay, 2000] showthat the energy Z, AU spectrumof the pickupprotons would dependon the solarwind turbulencelevel and may result in a relatively flat energy Figure 6. The axisymmetricheliospheric structure used in this distributionup to the 5-keV energies. This will also be work (calculationsby Baranovand Malama [1993, 1995]).The measurable. interstellarwind approachesfrom the left alongthe Z axis.BS, The energy-angleENA dependencesfor the two pickup bow shock;HP, heliopause;TS, terminationshock. protonevolution scenarios, as describedin section7, are shown in Plate 1 (middleand bottom panels). The middlepanel shows the pickup proton velocity distributionfunction in the solar the case of the pickup protons thermalized in the shock wind is a uniformlyfilled spherein the velocity spacewith the crossing.The pickupprotons are convectedinto the tail with the diameterequal to 2 Vsn,,where Vsv--450 km s" is thesolar wind temperaturescaled (cooled) with the local temperatureof the speedat 1 AU. This sphericaldistribution function crosses the bulk plasma. The bottom panel showsthe original spherical terminationshock unchanged.Beyond the shock,the pickup pickup proton distributionbeing convectedby the two-shock protonsare convectedby the plasmaflow field calculatedfor model flow field without thermalizationand without cooling. the two-shockgasdynamical model describedabove. In other Possiblecompression of pickup protonsin the shockcrossing words,no pickup protonheating and compressionat the shock wouldresult in slightlyhigher ENA fluxes. crossingoccurs, and the pickupprotons are not cooledas they The thermalizedpickup proton populationproduces ENA are convectedto the heliospherictail. It is assumedhere and in distributions,not unlike those for the stronglyshocked solar sections8 and 9 that pickupprotons constitute 15% of the solar wind plasmabut extendingto muchhigher energies because of wind at the termination shock. the significantlyhigher effective temperatureof the protons. Second,we assumethat the sameuniform spherical velocity The distributionof the ENAs producedby nonthermalized distribution is thermalized into a Maxwellian distribution in the pickupprotons (Plate 1, bottom)shows an abruptcutoff at high shocktransition; no energy exchangeoccurs with the bulk of energiesand strikingly different angular dependence. The cutoff the solar wind plasma. The pickup protons can now be energyin a given directionis determinedby the maximum describedby a shifted Maxwellian distributionfunction. In possiblevelocity of pickup protonsand by the plasmaradial other words, we assumeno heating and compressionat the velocityalong the line of sight.Characteristically, the ENA flux shock crossing,but the pickup proton energy is redistributed peaksin the downwindhemisphere in the 120ø < 0 < 150ø among the protonsto form a Maxwellian distribution.This range,which is an effectof the minimalradial plasma velocity Maxwellian pickup proton population is convectedto the and extendedENA-producing region in these directions.By heliospherictail. We assumethat the temperatureof the pickup measuringthe energy cutoff and absoluteENA fluxes, one protonschanges in proportionto the changeof the bulk plasma could determine the solar wind turbulence and its latitudinal temperature. dependence.The effectsof the solarwind turbulencewould be interwovenwith the possibleshock compression. 8. Imaging the Heliospherein ENA Fluxes Representativeall-sky ENA images in a selectedenergy The energy-angledependences of ENA fluxesare shownin band, 0.24 keV < E < 0.55 keV, are shown in Plate 2 in a Plate 1. The axesare the ENA energyand the observationangle conventionalMercator projectionin the ecliptic coordinates 0 (see Figure 1) measuredfrom the upwind direction.This with colorcoding of ENA fluxes.This and otherenergy bands format is convenientfor an axisymmetricheliosphere and are those of a realistic ENA single-pixelimager based on allowsdisplay of the whole sky for all ENA energies.The top availabletechnology and optimizedfor heliosphereimaging panel of Plate 1 shows the energy-angledependence of [Funstenet al., 1998b]. heliosphericENAs for a stronggasdynamical shock. For the gasdynamicalstrong shock case the ENA flux (cm'2 The ENA flux peaksin the upwinddirection. While ENAs s-• sr-•) peaksin the upwinddirection (ecliptic longitude 252 ø with <0.2 keV accountfor most of the flux, the wings of the and latitude+7 ø) for the interstellargas flow relativeto the Sun 15,774 GRUNTMAN ET AL.' ENERGETIC NEUTRAL ATOM IMAGING OF THE HELIOSPHERE

Hellospheric ENA Fluxes

0 1000 2000 3000 4000 5000 ENA flux, 1 / (crm2 sec sr keV)

0.40 i

0.30

0.20

z 0.10-

0.00

0 30 60 90 120 150 180

0.40

0.30

0.20

x • D c

z 0.10

0.00

0 30 60 90 120 150 180

1.2

0.9

,C x

0.0 i 0 30 60 90 120 ' 1 50 180 - ongle theto, degree Plate1. Energy-angleall-sky ENA images for (top) a stronggas-dynamical shock,(middle) asphericai isotropic pickupproton population with thermalization in theshock transition and cooling during convection to theheliospheric tail and(bottom) a sphericalisotropic pickup proton population without thermalization and without cooling, Note that the fluxesare multiplied by 1.0, 7.5, and2.5 for the top,middle, and bottom panels, respectively, to fit intothe same color scale. GRUNTMAN ET AL.: ENERGETIC NEUTRAL ATOM IMAGING OF THE HELlOSPHERE 15,775

Hellospheric ENA Fluxes 0.24 key <'E < 0.5.5 key

0 70 140 210 280 5.50 ENA flux, 1 / (crm2 sec st)

60-

30- o

-30

-6O

-9O

0 60 120 180 240 300 360

90 -•

30- o

¸ ' x

•-30 •_ o

-60

-90 • i 0 60 120 180 240 300 360

60 .o_

• • 30 o .

o x -- 0 m

• ::3 o ß? -50 •- '- ß

-60 "

-9O

o 60 120 180 240 300 560 ecliptic' longitude Plat•2. All-skyENA images inthe 0.24 keV < E< 0.55keV energy band for (top) astrong gas-dynamical shock, (middle)a sphericalisotropic pickup proton population with thermalization in the shocktransition and cooling while being convectedto the heliospherictail, and (bottom)a sphericalisotropic pickup proton population without thermalizationand without cooling. Note that the fluxesare multiplied by 5.0, 8.0, and 1.0 for the top,middle, and bottompanels, respectively, to fit into the samecolor scale. 15,776 GRUNTMANET AL.: ENERGETICNEUTRAL ATOM IMAG1NG OF THE HELlOSPHERE

A Ecliptic north 1

Sun / / / • nterstellar wind Spacecraft orbit direction

N B In-eclipticscan 1•4ø x 360ø swath m oneorbil) Crosswind • I t rs II

All-skyscan (entiresky covered D w win lwice/erbit)

•ount• C Iq /pix It

21100,

5150

I 60

307

75 Up ind Do n ind $

Plate3. (a) A genericsolar-cell-powered spacecraft in a heliocentricelliptical orbit in theecliptic plane with a 3-year period.The orbit perihelion and aphelion are 1.0 and 3.1 AU, respectively,and the line of apsidesisperpendicular to thedirection of theinterstellar wind flow. The spacecraft spin axis is pointedat theSun. One ENA sensoris pointed normallyto the spin axis, and the other is pointed parallel to the spin axis in theantisolar direction. (b) Covering of the celestialsphere by ENA sensorsin a "fish-eye"projection centered on the direction cross-wind tothe interstellar wind gasflow. (c) Total simulated ENA counts in each7øx7 ø pixelover the 0.24 keV

1 .E+04 hemisphereis a uniqueand highly promisingway to explorethe : I I propertiesand extension of the heliospherictail. The observed solar wind asymmetry [McComas et al., •' 1.E+03 2000a] would be prominentlydisplayed as higher energiesof the ENAs from the polar directions,corresponding to the higher solarwind velocities.The higher energiesare expectedfor a

z LU 1.E+00 . O= 122.5 deg I•_•, •, 9. Hellosphere Imaging Mission and Instrumentation i i I i I I I I I I I 1 .E-01 i i 0.00 0.25 0.50 0.75 1.00 1.25 1.50 The expectedENA fluxes are exceptionallyweak. It may ENA energy, keV takemore than a yearto obtainan imageof theheliosphere. An optimal approachseems to be to use a single-pixelENA Figure 7. Energydependence of ENA fluxesfor four different instrument with a large geometric factor to construct a directionsfor the strong-shockgasdynamical model. Angle 0 is compositeimage by scanningthe sky by combiningthe measuredfrom the upwinddirection. spacecraftspinning and evolution of the pointing of the spacecraftspin axis. Optimization of sucha complexapproach is not trivial andrequires careful planning and implementation. (Plate2, top). The gasdynamicalheating is strongestat the nose The feasibilitystudy of an experimentto imagethe heliosphere of the heliosphere.The shockedflow stagnatesin this regionof in ENA fluxes[Gruntman et al., 1999]was performed as a part the heliosphericsheath, before beginning to streamaround the of the developmentof the Interstellar Pathfinder mission terminationshock toward the tail. The radial componentof the [Gloeckleret al., 1999]. A realisticsingle-pixel heliospheric plasmavelocity is thusreduced to an outwardminimum. Under ENA sensor has a field of view-7øx7 ø and an effective suchconditions, the ENAs are generatedby the protonsin the geometricfactor 0.001 cm2 sr at 0.3 keV, 0.005 cm2 sr at 0.9 coreof the heatedphase space distribution and canbe observed keV,and 0.020 cm 2 sr at 6.0 keV [Funstenet al., 1998b].The at 1 AU at essentiallythe same energy as that their parent energyrange of interest,0.2-6.0 keV, is coveredby six protonshad in the plasmaframe. contiguousenergy bands, with the energychannel limits 0.24, An ENA image of the thermalized pickup protons is 0.55, 1.2, 2.3, 3.8, and5.9 keV, asshown in Figure9. expectedlysimilar to that of the strong-shockcase but extends The spacecraftshould be well outsidethe magnetosphere, to muchhigher energies (Plate 2, middle). The secondarypeak whichis a copioussource of ENAs at all energiesof interest between0 = 120ø and 180ø is an effect of the plasmaflow here. A spinning spacecraftis most desirablewith the ENA around the heliospherewith the extended ENA-producing instrumentboresight pointed normally to thespin axis, to sweep regiondownwind. The ENA fluxesexhibit an entirelydifferent a swathin the sky with eachspin. If the spinaxis precesses image (Plate 2, bottom) in the case of the sphericalvelocity aboutan axis perpendicularto the spin axis over a periodT, distribution of pickup protons without thermalization and then the entire celestialsphere will be scannedin a half of that withoutcooling. The fluxespeak in the downwindhemisphere time, T/2. in the shapeof a broadannulus 70o-80 ø in diameter. A convenientway to achieve the desired experiment Figure 8 showsthe ENA flux angular dependencesin two geometryis to point the spacecraftspin axis at the Sun. A energybands (0.24

• 1000.0 For the 1x3.1 AU orbit the greatlyreduced angular velocity t: I I I ! of the spacecraftnear aphelionproduces the high exposure times(70 daysper 7øx7ø pixel) for the in-eclipticimager in the 0.24keY< E<0.55 keV.J cross-winddirection. The same orbital dynamicenhances the exposuretimes for the spinningall-sky imagerin the upwind .lOO.O and downwind directions, relative to a 1-AU circular orbit. Thusthe 1x3.1 AU orbit actuallyincreases the sensitivityof the i all-sky imager to the expectedimportant upwind/downwind < ...... I1...... asymmetriesin ENA fluxes. Similarly,the very high exposure o 10.0 time for the in-ecliptichead in the cross-winddirection opens . the possibilityof detectingtemporal variations on the ecliptic flanks of the termination shock. Plate 3c demonstrates total simulated ENA counts in the 0.24

LU 1.0 i keV

1000 duty cycle achievedfor the polar heliosphericregions and ...•. Instrumentenergychannels minimalduty cycleachieved for the regionsclose to the ecliptic plane.The directionaldependence of the coverageduty cycle is •100 compensatedfor during the data reduction. Since both Pickupprotons -: instrumentswould cover some identical directionsin the sky Gasdynamical' duringthe mission, they can be intercalibrated. '•• 10 r--'-•Y------ß upwind -.. The transverse(spinning) imager will cover the entire • '--J : celestialsphere in half of the orbitperiod, T/2, but with a low Gasdynamical2 downwind - duty cycle. On the otherhand, the parallelimager will only • 0 • I! ,•, ,, I , sweepthe eclipticdirections in the periodT, but with -50 times 0 I 2 3 4 5 6 the duty cycleof the spinningimager. Much highernumbers of Energy,keV counts accumulated in ecliptic pixels are important for identifyingpossible contributions from nonterminationshock Figure 9. Calculatedinstrument ENA countrates for a strong ENA sourcesinside the heliosphere.This ENA foreground (gasdynamical)shock in the upwind and downwinddirections includes contributionsfrom the energeticproton populations and for pickup protons(model of Fahr and Lay [2000]). The associatedwith shocksin the solarwind [Roelof,1992]. instrumentenergy channels are shownat the top of Figure9. GRUNTMAN ET AL.: ENERGETIC NEUTRAL ATOM IMAGING OF THE HELIOSPHERE 15,779

The measurementtechnique is basedon ionizing (stripping) Chalov,S.V., and H.J. Fahr,Reflection of preacceleratedpickup ions ENAs by transmissionthrough ultra-thin foils followedby the at the solar wind termination shock: The seed for anomalous energyanalysis of the ionizedENAs. The lower energylimit for cosmicrays, Sol. Phys., 168, 389-411, 1996. Chalov,S.V., andH.J. Fahr,Pick-up ion accelerationat the termination the_technique is determined by the strippingefficiency of the shock and post-shockpick-up ion energy distribution,Astron. foils. The techniqueefficiently works down to the energiesas Astrophys.,360, 381-390, 2000. low as 200-250 eV [Overburyet al., 1979; Funstenet al., 1994, Chalov,S.V., H.J. Fahr, and V. Izmodenov,Spectra of energizedpick- 1998a, 1998b]. up ions upstreamof the 2-dimensionalheliospheric termination shock,1, The role of Alfvenicturbulences, Astron. Astrophys., 304, To conclude,ENA imaging is a powerful techniqueto 609-616, 1995. remotelyprobe properties of distanthot plasmas.The technique Chalov,S.V., H.J. Fahr,and V. Izmodenov,Spectra of energizedpick- ideally suits the study of the asynmaetric,three-dimensional up ions upstreamof the 2-dimensionalheliospheric termination heliosphericinterface. ENA imaging in the 0.2-6 keV energy shock,2, Accelerationby Alfvenic and by large-scalesolar wind range will establishthe nature of the terminationshock and turbulences,Astron. Astrophys., 320, 659-671, 1997. Cox, D.P., and R.J. Reynolds, The local interstellarmedium, Annu. propertiesof hot proton populationsin the heliosheath.The Rev.Astron. Astrophys., 25, 303-344, 1987. evolutionof pickupproton population at andbeyond the shock Czechowski,A., S. Grzedzielski,and I. Mostafa, Apex-antiapex can be remotely explored.Heliosphere ENA imaging would asymmetry in the anomalouscosmic ray distribution in the distinguishamong the competingSun-LISM interactionmodels heliosheath,Astron. Astrophys., 297, 892-898, 1995. Davis, L., Jr., Interplanetarymagnetic fields and cosmicrays, Phys. and explorethe physicsof the criticalprocesses in the interface Rev., 100, 1440-1444, 1955. region. Weak ENA fluxes make imaging implementation Dessler,A. J., Solar wind and interplanetarymagnetic field, Rev. nontrivial,but it canbe done,as we show,by single-pixelENA Geophys.,5, 1-41, 1967. sensorson a spinningspacecraft during a typicalmission near 1 Fahr, H. J., On the influenceof neutralinterstellar matter on the upper AU. The instrumentationtechnology is ready. atmosphere,Astrophys. Space $ci., 2, 474-495, 1968a. Fahr, H. J., Neutral corpuscularenergy flux by charge-transfer Acknowledgements.The authorsthank Vladimir Baranovand his collisionsin the vicinity of the Sun,Astrophys. Space $ci., 2, 496- group for calculatingthe heliosphericplasma-gas structure for this 503, 1968b. work. The effortsof E. C. R. and D. G. M. were supportedin part by Fahr,H.J., Non-thermalsolar wind heatingby suprathermalions, Sol. grantNAG5-3879 from NASA to the JohnsHopkins University. Work Phys.,30, 193-206,1973. at Los Alamos National Laboratory (H.O.F. and D.J.M.) was Fahr, H. J., The extraterrestrialUV-background and the nearby performedunder auspicesof the U.S. Departmentof Energy. M.G. interstellarmedium, Space $ci. Rev., 15, 483-540, 1974. was supportedin part by NASA grantNAG5-6966. Fahr,H.J., The interstellargas flow throughthe heliosphericinterface Janet G. Luhmann thanks Earl Scime and another referee for their region,Space $ci. Rev., 78, 199-212, 1996. assistancein evaluatingthis paper. Fahr,H.J., and J. Ziemkiewicz,The behaviorof distantheliospheric pick-upions and associated solar wind heating, Astron. Astrophys., 202, 295-305, 1988. References Fahr, H.J., and H. Fichtner,Physical reasons and consequencesof a three-dimensionallystructured heliosphere, Space $ci. Rev., 58, Axford, W. I., The interaction of the solar wind with the interstellar 193-258, 1991. medium,in Solar Wind,NASA Spec. Publ., SP-308, 609-660, 1972. Fahr,H.J., and D. Rucinski,Neutral interstellar gas atoms reducing the Axford, W. I., A. J. Dessler, and B. Gottlieb, Termination of solar solar wind Mach numberand fractionallyneutralizing the solar wind andsolar magnetic field, Astrophys. d., 137, 1268-1278,1963. wind, Astron.Astrophys., 350, 1071-1078, 1999. Baranov, V. B., Gasdynamicsof the solar wind interactionwith the Fahr, H.J., and G. Lay, Remote diagnosticof the heliospheric interstellarmedium, Space $ci. Rev.,52, 89-120, 1990. terminationshock using neutralizedpost-shock pick-up ions as Baranov, V. B., and Y. G. Malama, The model of the solar wind messengers,Astron. Astrophys., 356, 327-334, 2000. interaction with the local interstellar medium: Numerical solution Fahr, H.J., R. Osterbart,and D. Rucinski,Modulation of the interstellar of self-consistentproblem, d. Geophys.Res., 98, 15,157-15,163, oxygen-to-hydrogenratio by the heliosphericinterface plasma, 1993. Astron.Astrophys., 294, 587-600, 1995. Baranov, V. B., and Y. G. Malama, Effect of local interstellar medium Fahr, H.J., T. Kausch,and H. Scherer,A 5-fluid hydrodynamic hydrogen fractional ionization on the distant solar wind and approachto modelthe solarsystem-interstellar medium interaction, interfaceregion, d. Geophys.Res., 100, 14,755-14,761, 1995. Astron.Astrophys., 35 7, 268-282, 2000. Baranov, V.B., V.V. Izmodenov, and Y.G. Malama, On the Fisk,L.A., Implicationsof a weaktermination shock, Space $ci. Rev., distributionfunction of H-atoms in the problemof the solarwind 78, 129-136, 1996. interactionwith the local interstellarmedium, d. Geophys.Res., Fisk, L.A., B. Kozlovsky,and R. Ramaty,An interpretationof the 103, 9575-9585, 1998. observedoxygen and nitrogen enhancements in low-energy cosmic Bertaux,J.-L., Helium and hydrogenof the local interstellarmedium rays, Astrophys.d., 190, L35-L37, 1974. observedin the vicinity of the Sun, in IAU ColloquiumNo. 81, Frisch,P.C., Characteristicsof nearbyinterstellar matter, Space $ci. NASAConf Publ., CP 2345, 3-23, 1984. Rev., 72, 499-592, 1995. Bertaux,J.L., E. Quemerais,and R. Lallement,Observation of a sky Funsten,H.O., D.J. McComas,and E.E. Scime,Comparative study of Lyman tx grooverelated to enhancedsolar wind mass flux in the low energyneutral atom imaging techniques, Opt. Eng., 33, 349- neutralsheet, Geophys. Res. Lett., 23, 3675-3678, 1996. 356, 1994. Bertaux,J.-L., E. Kyr61•i,E. Quemerais,R. Lallement,W. Schmidt,T. Funsten,H.O., D.J. McComas,and E.E. Scime,Low-energy neutral- Summanen, J. Costa, and T. M•ikinen, SWAN observationsof the atom imaging techniques for remote observations of the solar wind latitude distribution and its evolution since launch, magnetosphere,d. $pacecr.Rockets, 32, 899-904, 1995. Space$ci. Rev., 87, 129-132, 1999. Funsten, H.O., D.J. McComas, and M.A. Gruntman, Neutral atom Bogdan,T.J., M.A. Lee, and P. Schneider,Coupled quasi-linear wave imaging:UV rejectiontechniques, in MeasurementTechniques in dampingand stochasticacceleration of pickup ions in the solar SpacePlasmas.' Fields, Geophys.Monogr. Set., Vol.103, editedby wind, d. Geophys.Res., 96, 161-178, 1991. R. Pfaff, J. Borovsky, and D.T. Young, pp. 251-256, AGU, Brandt,P. C., S. Barabash,O. Norberg, R. Lundin, E.C. Roelof, and Washington,1998a. C.J. Chase,Energetic neutral atom imaging at low altitudesfrom Funsten,H.O., D.J. McComas, D.G. Mitchell, E. C. Roelof, and M.A. the Swedishmicrosatellite Astrid: Imagesand spectralanalysis, d. Gruntman,Energetic neutral atoms imaging of the heliospheric Geophys.Res.,104, 2367-2380, 1999. terminationshock, Eos Trans.AGU, 79(45), Fall Meet. Suppl., Breitschwerdt,D., The local bubble, Space $ci. Rev., 78, 173-182, F705, 1998b. 1996. Gloeckler,G., J. Geiss,H. Balsiger,L. A. Fisk, A. B. Galvin, F. M. Chalov, S.V., and H.J. Fahr, A two-fluid model of the solar wind Ipavich,K. W. Ogilvie,R. von Steiger,and B. Wilken, Detectionof termination shock modified by shock-generatedcosmic rays interstellarpickup hydrogen in the solarsystem, Science, 261, 70- includingenergy losses, Astron. Astrophys., 228, 973-980, 1994. 73, 1993. 15,780 GRUNTMAN ET AL.: ENERGETIC NEUTRAL ATOM IMAGING OF THE HELIOSPHERE

Gloeckler, G., J. Geiss, E.C. Roelof, L.A. Fisk, F.M. Ipavich, K.W. Krimigis,S. M., et al., MagnetosphereImaging Instrument (MIMI) on Ogilvie, L.J. Lanzerotti, R. von Steiger, and B. Wilken, the CassiniMission to Saturn/Titan,Space Sci. Rev., in press,2001. Accelerationof interstellarpickup ions in the disturbedsolar wind Lallement,R., J.-L. Bertaux,E. Chassefiere,and B.R. Sandel,Lyman- observedon Ulysses,J. Geophys.Res., 99, 17,637-17,643,1994. alpha observationsfrom Voyager (1-18 AU), in Physicsof the Gloeckler,G., E.C. Roelof, K.W. Ogilvie, and D.B. Berdichevsky, OuterHeliosphere, edited by S. Grzedzielskiand D.E. Page,pp.73- Proton phase space densities(0.5 eV < Ep < 5 MeV) at mid- 82, Pergamon,Tarrytown, N.Y., 1990. latitudes from Ulysses SWICS/HI-SCALE measurements,Space Lallement, R., J.-L. Bertaux, and J.T. Clarke, Deceleration of Sci. Rev., 72, 321-326, 1995. interstellarhydrogen at the hellosphericinterface, Science, 260, Gloeckler, G., L.A. Fisk, T.H. Zurbuchen,E. Moebius, H.O. Funsten, 1095-1098, 1993. M. Witte, and E.C. Roelof, InterstellarPathfinder: A missionto the Lee, M.A., The terminationshock of the solar wind, SpaceSci. Rev., inner edge of the interstellarmedium, Eos Trans.,4GU, 80(17), 78, 109-116, 1996. SpringMeet. Suppl.,S237, 1999. LeRoux, J.A. and H. Fichtner,The influenceof pickup, anomalous, Gruntman,M. A., Effect of neutral componentof solar wind on the and galacticcosmic-ray protons on the hellosphericshock: A self- interactionof the solar systemwith the interstellargas flow, Sov. consistentapproach, Astrophys. J. Lett., 477, L115-L118, 1997. ,4stroh.Lett., 8(1), 24-26, 1982. Linde, T.J., T.I. Gombosi,P.L. Roe, K.G. Powell, and D.L. DeZeeuw, Gruntman,M. A., Anisotropyof the energeticneutral atom flux in the Hellospherein themagnetized local interstellar medium: Results of heliosphere,Planet. Space Sci., 40, 439-445, 1992. a 3-dimensionalMHD simulation,J. Geophys.Res., 103, 1889- Gruntman,M. A., Neutralsolar wind properties:Advance warning of 1904, 1998. majorgeomagnetic storms, J. Geophys.Res., 99, 19,213-19,227, Linsky, J.L., S. Redfield, B.E. Wood, and N. Piskunov,The three- 1994. dimensional structure of the warm local interstellar medium, I, Gruntman,M., Energeticneutral atom imaging of spaceplasmas, Rev. Methodology,Astrophys. J., 528, 756-766, 2000. Sci. Instrum., 68, 3617-3656, 1997. McComas,D.J., H.O. Funsten,J.T. Gosling,K.R. Moore, E.E. Scime, Gruntman,M., Imagingthe three-dimensionalsolar wind, J. Geophys. and M.F. Thomsen, Fundamentalsof low-energy neutral atom Res., 106. 8205-8216, 2001 a. imaging,Opt. Eng., 33, 335-341, 1994. Gruntman,M., Mappingthe heliopause in EUV, in Proceedingsof the McComas,D.J., H.O. Funsten,J.T. Gosling,and W.R. Pryor, Ulysses COSPAR Colloquiumon the Physicsof the Outer Hellosphere, measurementsof variationsin the solarwind-interstellar hydrogen ElsevierSci., New York, in press,2001 b. chargeexchange rate, Geophys.Res. Lett., 26, 2701-2704, 1999. Gruntman,M., and H.J. Fahr, Accessto the heliosphericboundary: McComas, D.J., B.L. Barraclough,H.O. Funsten,J.T. Gosling, E. EUV-echoes from the hellopause,Geophys. Res. Lett., 25, 1261- Santiago-Mufioz,R.M. Skoug,B.E. Goldstein,M. Neugebauer,P. 1264, 1998. Riley, andA. Balogh,Solar wind observationsover Ulysses'first Gruntman,M., andH.J. Fahr,Heliopause imaging in EUV: OxygenO + full polar orbit, J. Geophys.Res., 105, 10,419-10,433,2000a. ion 83.4-nm resonanceline emission,d. Geophys.Res., 105, 5189- McComas,D.J., J.T. Gosling, and R.M. Skoug,Ulysses observations 5200, 2000. of the irregularly structuredmid-latitude solar wind during the Gruntman,M.A., V.B. Leonas,and S. Grzedzielski,Neutral solarwind approachto solar minimum, Geophys.Res. Lett., 27, 2437-2440, experiment,in Physicsof the Outer Heliosphere,edited by S. 2000b. Grzedzielski and D.E. Page, pp..355-358, Pergamon,Tarrytown, Meier, R. R., Some optical and kinetic propertiesof the nearby N.Y., 1990. interstellargas, Astron.Astrophys., 55, 211-219, 1977. Gruntman, M.A., E.C. Roelof, D.G. Mitchell, H.O. Funsten, D.J. Mitchell, D. G., et al., INCA: The ion neutralcamera for energetic McComas,and H.-J. Fahr, Energeticneutral atom imagingof the neutralatom imagingof the Saturnianmagnetosphere, Opt. Eng., heliospherictermination shock and interface region on the 32, 3096-3101, 1993. "InterstellarPathfinder" mission, Eos Trans.A GU, 80(17), Spring Mitchell, D.G., et al., High energyneutral atom (HENA) imagerfor Meet. Suppl.,S237, 1999. the IMAGE mission,Space Sci. Rev., 91, 67-112, 2000. Henderson,M.G., G.D. Reeves, H.E. Spence,R.B. Sheldon, A.M. Moebius, E., D. Hovestadt,B. -Klecker,M. Scholer,G. Gloeckler, and Jorgensen,J.B. Blake,and J.F. Fennel,First energetic neutral atom F.M. Ipavich, Direct observationof He* pickup ions of interstellar imagesfrom Polar, Geophys.Res. Lett., 24, 1167-1170,1997. origin in the solarwind, Nature, 318, 426-429, 1985. Hilchenbach,M., et al., Detection of 55-80 keV hydrogenatoms of Moore, T.E., et al., Low energy neutral atom (LENA) imager for heliosphericorigin by CELIAS/HSTOF on SOHO, Astrophys.J., IMAGE, SpaceSci. Rev., 91,155-195, 2000. 503, 916-922, 1998. Morton, D.C., and J. D. Purcell, Observations of the extreme Holzer, T. E., Neutral hydrogen in interplanetaryspace, Rev. ultravioletradiation in the night sky using an atomic hydrogen Geophys..,15, 467-490, 1977. filter, Planet. SpaceSci., 9, 455-458, 1962. Holzer, T.E., and W.I. Axford, Solar wind ion composition,J. Neugebauer,M., The three-dimensionalsolar wind at solar activity Geophys.Res., 75, 6354-6359, 1970. minimum,Rev. Geophys.,37, 107-126, 1999. Hsieh, K. C., and M. A. Gruntman,Viewing the outerheliosphere in Orsini,S., et al., Proposalof an Italianexperiment for the mission energeticneutral atoms, Adv. SpaceRes., 13(6), 131-139, 1993. SAC-B. ISENA: Imaging particle spectrometerfor energetic Hsieh,K. C., K. L. Shih, J. R. Jokipii,and S. Grzedzielski,Probing the neutral atoms, in Instrumentationfor MagnetosphericImagery, hellospherewith energetic neutral atoms, Astrophys.J., 393, editedby S. Chakrabarti,Proc. SPIE Int. Soc. Opt. Eng., 1744, 91- 756-763, 1992. 101, 1992 Isenberg,P.A., Interactionof the solar wind with interstellarneutral Osterbart,R., andH.J. Fahr, A Boltzmann-kineticapproach to describe hydrogen:Three-fluid model, J. Geophys.Res., 91, 9965-9972, the entranceof neutral interstellarhydrogen into the heliosphere, 1986. Astron.Astrophys., 264, 260-269, 1992. Isenberg, P.A., A hemisphericalmodel of anisotropicinterstellar Overbury,S. H., P. F. Dittner, S. Datz, and R. S. Thoe, Energyloss, pickupions, J. Geophys.Res., 102, 4719-4724, 1997a. angulardistributions and chargefractions of low energyhydrogen Isenberg,P.A., A weakersolar wind terminationshock, Geophys. Res. transmittedthrough thin carbonfoils, Radiat. Eft, 41, 219-227, Lett., 24, 623-626, 1997b. 1979. Izmodenov, V., Y.G. Malama, and R. Lallement, Interstellar neutral Parker,E.N., InterplanetaryDynamical Processes, Wiley-Interscience, oxygenin a two-shockheliosphere, Astron. Astrophys., 317, 193- New York, 1963. 202, 1997. Patterson,T. N. L., F. S. Johnson,and W. B. Hanson, The distribution Jaeger,S., andH.J. Fahr, The hellosphericplasma tail underinfluence of interplanetaryhydrogen, Planet. SpaceSci., 11,767-778, 1963. of chargeexchange processes with interstellarH-atoms, Sol. Phys., Pesses,M.E., J.R. Jokipii, and D. Eichler, Cosmic ray drift, shock 178, 631-656, 1998. wave acceleration,and the anomalouscomponent of cosmicrays, Jokipii,J.R., and J. Giacalone,The accelerationof pickupions, Space Astrophys.J. Lett., 246, L85-L88, 1981. Sci. Rev., 78, 137-148, 1996. Pollock,C.J., et al., Medium energyneutral atom (MENA) imagerfor Keppler,E., M. Franz,A. Korth,M.K. Reuss,J.B. Blake, R. Seidel,J.J. the IMAGE mission,Space Sci. Rev., 91, 113-154, 2000. Quenby, and M. Witte, Observationsof energeticparticles with Ratkiewicz,R., A. Barnes,G.A. Molvik, J.R. Spreiter,S.S. Stahara,M. EPAC on Ulyssesin polar latitudeof the heliosphere,Science, 268, Vinokur, and S. Venkateswaran,Effect of varying strengthand 1013-1016, 1995. orientationof local interstellarmagnetic field on configurationof GRUNTMAN ET AL.: ENERGETIC NEUTRAL ATOM IMAGING OF THE HELlOSPHERE 15,781

exteriorheliosphere: 3D MHD simulations,Astron. Astrophys., Wood, B.E., and J.L. Linsky, The local ISM and its interactionwith 335, 363-369, 1998. the winds of nearby late-type stars,Astrophys. J., 492, 788-803, Ripken,H.W., and H.J. Fahr,Modification of the localinterstellar gas 1998. propertiesin the heliosphericinterface, Astron. Astrophys., 122, Yoon, P.H., L.F. Ziebell, and C.S. Wu, Self-consistentpitch angle 181-192, 1983. diffusionof newbornions, J. Geophys.Res., 96, 5469-5478, 1991. Roelof, E. C., Energeticneutral atom image of a storm-timering Zank, G.P, Interaction of the solar wind with the local interstellar current,Geophys. Res. Lett., 14, 652-655, 1987. medium: A theoreticalperspective, Space Sci. Rev., 89, 1-275, Roelof, E. C., Imaging heliosphericshocks using energeticneutral 1999. atoms, in Solar WindSeven, edited by E. Marschand R. Schwenn, Zank, G.P., G.M. Web, and D.J. Donohue,Particle injection and the pp.385-394,Pergamon, Tarrytown, N.Y., 1992. structureof energetic-particle-modifiedshocks, Astrophys. J., 406, Roelof,E.C., D.G. Mitchell, and S.M. Krimigis,ENA imagesfrom the 67-91, 1993. ion andneutral camera (INCA) duringthe CassiniEarth flyby, Eos Zank, G.P., H.L. Pauls, I.H. Cairns, and G.M. Webb, Interstellar Trans.AGU, 80(46), Fall Meet. Suppl.,F872, 1999. pickup ions and quasi-perpendicularshocks: Implications for the Rucinski,D., A.C. Cummings,G. Gloeckler,A.J. Lazarus,E. M6bius, terminationshock and interplanetaryshocks, J. Geophys.Res., 101, and M. Witte, Ionizationprocesses in the heliosphere:Rates and 457-477, 1996. methodsof their determination,Space Sci. Rev., 78, 73-84, 1996. Suess,S.T., The heliopause,Rev. Geophys.,28, 97-115, 1990. H.J. Fahr, Institut far Astrophysik und Extraterrestrische Suess,S.T., Temporalvariations in the terminationshock distance, J. Forschung,Universit•tt Bonn, BonnD-53121, Germany. Geophys.Res., 98, 15,147-15,155,1993. H.O. Funsten,Los Alamos National Laboratory,Los Alamos, NM Thomas, G. E., The interstellar wind and its influence on the 87545. interplanetaryenvironment, Annu. Rev. Earth Planet. Sci., 6, 173- M. Gruntman, Department of Aerospace and Mechanical 204, 1978. Engineering, MC-1191, University of Southern California, Los Vasyliunas,V.M., and G.L. Siscoe, On the flux and the energy Angeles,CA 90089-1191. (mikeg•spock.usc.edu) spectrumof interstellarions in the solar system,J. Geophys.Res., D.J. McComas, Southwest Research Institute, San Antonio, TX 81, 1247-1252, 1976. 78228-0510. Wallis, M., Local interstellarmedium, Nature, 254, 202-203, 1975. D.G. Mitchelland E.C. Roelof,Applied Physics Laboratory, Johns Williams, D. J., E. C. Roelof, and D. G. Mitchell, Global HopkinsUniversity, Laurel, MD 20707-6099. magnetosphereimaging, Rev. Geophys.,30, 183-208, 1992. Witte, M., M. Banaszkiewicz, and H. Rosenbauer, Recent results on the parametersof the interstellarhelium from the Ulysses/GAS (ReceivedAugust 27, 2000; revisedOctober 16, 2000; experiment,Space Sci. Rev., 78, 289-296, 1996. acceptedOctober 23, 2000.)