methodologies and techniques for detecting extraterrestrial...

ASTROBIOLOGYVolume 3 Number 3 2003copy Mary Ann Liebert Inc

Methodologies and Techniques for DetectingExtraterrestrial (Microbial) Life

Near-Infrared Detection of Potential Evidence forMicroscopic Organisms on Europa

J BRAD DALTON1 RAKESH MOGUL2 HIROMI K KAGAWA1 SUZANNE L CHAN1

and COREY S JAMIESON3

ABSTRACT

The possibility of an ocean within the icy shell of Jupiterrsquos moon Europa has established thatworld as a primary candidate in the search for extraterrestrial life within our Solar SystemThis paper evaluates the potential to detect evidence for microbial life by comparing labora-tory studies of terrestrial microorganisms with measurements from the Galileo Near InfraredImaging Spectrometer (NIMS) If the interior of Europa at one time harbored life some evi-dence may remain in the surface materials Examination of laboratory spectra of terrestrialextremophiles measured at cryogenic temperatures reveals distorted asymmetric near-infrared absorption features due to water of hydration The band centers widths and shapesof these features closely match those observed in the Europa spectra These features arestrongest in reddish-brown disrupted terrains such as linea and chaos regions Narrow spec-tral features due to amide bonds in the microbe proteins provide a means of constraining theabundances of such materials using the NIMS data The NIMS data of disrupted terrains ex-hibit distorted asymmetric near-infrared absorption features consistent with the presence ofwater ice sulfuric acid octahydrate hydrated salts and possibly as much as 02 mg cm23 ofcarbonaceous material that could be of biological origin However inherent noise in the ob-servations and limitations of spectral sampling must be taken into account when discussingthese findings Key Words Extraterrestrial lifemdashEuropamdashInfrared spectroscopymdashExtremo-philes Astrobiology 3 505ndash529

505

INTRODUCTION

EUROPA HAS BEEN IDENTIFIED as a potential hostfor extraterrestrial life within our Solar Sys-

tem Recent work has indicated the likelihood ofnot only a possible subsurface ocean (Cassen et

al 1979 Pappalardo et al 1999 Kivelson et al2000 Stevenson 2000) but also a significant com-plement of the biogenic elements (Pierazzo andChyba 2002) It has been postulated that thisocean could have at one time harbored prebioticcompounds or even extraterrestrial life (Reynolds

1SETI Institute Mountain View California2Wilkes Honor College Florida Atlantic University Jupiter Florida3University of Hawaii Manoa Hawaii

et al 1983 Sagan et al 1993 Fanale et al 1998Chyba and Phillips 2001 Rothschild and Man-cinelli 2001 Zolotov and Shock 2003)

Because the putative ocean remains inaccessi-ble beneath Europarsquos icy crust infrared remotesensing of the surface materials has been used toinfer details about the interior composition If infact the surface composition is indicative of inte-rior composition studies of surface materialsmay provide insight into biological and prebioticprocesses (Squyres et al 1983 McCord et al1999a Kargel et al 2000) The diagnostic spectralcharacteristics of specific biogenic materials mayprovide unique biosignatures (Schulze-Makuchet al 2002) which can be tested for in the searchfor extraterrestrial life Infrared detection of evi-dence for life requires that (1) material of biolog-ical origin be emplaced at the surface (2) identi-fiable infrared absorption features diagnostic ofthis material survive exposure to the hostile spaceenvironment and (3) spectra of sufficient spectralresolution and signal-to-noise ratio be acquiredat spatial scales capable of isolating regions con-taining the material of interest This paper appliesthese criteria to the examination of recent space-craft and laboratory data concerning the spectraof Europa and those of terrestrial microorgan-isms

The surface of Europa is composed mostly ofwater ice (Pilcher et al 1972 Clark and McCord1980) More recent ground-based and spacecraftobservations have indicated the presence of othermaterials as well Cometary and meteoric infallare expected to provide several elemental andmolecular species over the age of the Solar Sys-tem including biogenic elements (Zahnle et al1998 Pierazzo and Chyba 2002) Energetic parti-cle bombardment associated with Jupiterrsquos mag-netosphere not only delivers elements such as HS and O but also drives complex chemistry thatis expected to produce a number of additionalcompounds (Delitsky and Lane 1997 1998Cooper et al 2001 Carlson et al 2002) Observedspecies include O2 SO2 CO2 and H2O2 (Lane etal 1981 Noll et al 1995 Carlson et al 1999ab)Current models suggest the presence of sulfuricacid octahydrate and other sulfur compounds(Carlson et al 1999b) simple organics such asformaldehyde (Delitsky and Lane 1998 Chyba2000) and hydrated sulfate and carbonate salts(Kargel 1991 1998 McCord et al 1998 1999ab2002 Kargel et al 2000) While no single mater-ial completely explains the Galileo observations

diagnostic infrared absorption features haveproven useful for constraining the abundances ofseveral compounds (Dalton 2000)

Water ice and frost have well-characterizedsymmetric near-infrared absorption features at15 and 20 mm (Ockman 1957 Herzberg 1991Gaffey et al 1993) These are evident in theGalileo Near-Infrared Imaging Spectrometer(NIMS) Europa spectrum of icy terrain in Fig 1(bottom curve) Other regions of the surface dis-play distorted asymmetric features at these po-sitions as indicated by the top curve The pres-ence of these distorted absorption features intelescopic and spacecraft observations of Europa(Clark and McCord 1980 Dalton 2000) forms thebasis of the argument for hydrated salts on thesurface (McCord et al 1998 1999a) These fea-tures are clearly due to water of hydration (Paul-ing 1935) though the host molecule is less cer-tain (Gaffey et al 1993 Carlson et al 1999b

DALTON ET AL506

0

01

02

03

04

05

06

07

08

08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Icy Plains

Dark terrain

FIG 1 Galileo NIMS spectra of both icy and dark sur-face units on Europa Lower curve is a 26-NIMSEL av-erage from an icy region illustrating the broad 20-mmabsorption feature indicating pure H2O ice Upper curveis an average of 18 NIMSELS corresponding to dark red-dish-brown terrain This curve exhibits the distorted andasymmetric 20-mm absorption bands typical of boundwater such as water of hydration Both averages were ex-tracted from contiguous portions of the G1ENNHILAT01observation (Dalton 2000) Overlapping data segmentsat every 12th wavelength position are instrumental arti-facts created by stepping the grating to acquire readingsat each detector the resulting segments are compiled tocreate each spectrum (Carlson et al 1992) with redun-dancy supplied by the overlapping wavelengths

McCord et al 1999b 2002 Dalton 2003) Mosthydrated materials exhibit features at these posi-tions and the hydrated sulfate salts are only oneclass of minerals with these characteristics (Huntet al 1971 Crowley 1991 Herzberg 1991 Dal-ton 2000) In fact the hydrate features in the spec-tra of the sulfate salts are not intrinsic to the sul-fates but rather to the bound water itself andassociated hydrogen bond energy shifts (Hunt etal 1971 Gaffey et al 1993 Dalton and Clark1998 Chaban et al 2002) Comparison of GalileoNIMS spectra with Solid State Imager (SSI) dataindicates that the hydrate absorption features aremost pronounced in reddish-brown disruptedterrain units such as linea and chaos regions andin certain impact craters (Granahan et al 1998McCord et al 1998 Fanale et al 1999 2000) yetvirtually nonexistent in the bright plains unitsThis has been interpreted as evidence of a link to subsurface processes and interior composition(Kargel 1991 Kargel et al 2000 Spaun and Head2001 Zolotov and Shock 2001 McKinnon andShock 2002) though some role is probablyplayed by exogenic processes such as ion im-plantation and radiolysis (Carlson et al 1999b2002 Cooper et al 2001 Paranicas et al 2001)

The exact mechanisms of disruption and em-placement of these materials are the subject ofconsiderable debate (Squyres et al 1983 Spaunet al 1998 Greenberg et al 1999 Head and Pap-palardo 1999 Prockter and Pappalardo 2000Figueredo et al 2002 Prockter et al 2002) Thequestion of whether these surface units could beindicative of oceanic composition remains openIf there is no connection then investigation of as-trobiological potential at Europa will require ac-cess to the subsurface For the purposes of thisanalysis we shall assume that such a link existsIn that case if the ocean at one time containedsome form of organisms then it may be inferredthat these would be emplaced at the surface aswell

Once emplaced at the surface any materialwould be subjected to the hostile space environ-ment Energetic charged particle bombardmentultraviolet photolysis dehydration and chemicalbreakdown by radical species would all act to de-compose exposed materials (Cooper et al 2001Hudson and Moore 2001 Paranicas et al 2001Carlson et al 2002) To be detected biological orprebiotic material would have to persist to thepresent While the long-term viability of severalcandidate materials has been questioned (Dalton

and Clark 1998 Marion and Farren 1999 Zolo-tov and Shock 2001) recent laboratory and the-oretical work indicates that some compoundsmay be stable on geologic timescales (McCord etal 2001 Carlson et al 2002 Orlando 2002 Pier-azzo and Chyba 2002 Mogul et al 2003) Becausereaction rates depend strongly upon temperatureexperiments must be performed at the low tem-peratures relevant to Europarsquos surface (80ndash130 K)Several of the hydrated salts have been postu-lated to exhibit resistance to degradation over thelifetime of Europarsquos surface (McCord et al 2001)though Carlson et al (2002) predict destruction ofmany sulfate compounds in less than 4000 yearsIt has also been recently shown that survival ratesfor microorganisms and their component materi-als exposed to energetic particle bombardmentare dependent not only on energy but also uponcomposition of bombarding ions (Mogul et al2003) Extrapolation from room temperature ex-periments or experiments using particles and en-ergies other than those present at Europa may notbe appropriate As there is presently much dis-agreement in the literature regarding destructionrates and survival lifetimes for complex com-pounds in Europarsquos radiation environment fur-ther work will be needed to constrain the rates ofdestruction for all candidate materials

Because the infrared spectral features of manycompounds including water ice and water of hy-dration are strongly temperature-dependent in-terpretation of icy satellite spectra must rely onlaboratory measurements performed at or nearthe surface temperatures of these worlds (Finkand Sill 1982 Gaffey et al 1993 Grundy andSchmitt 1998 Roush 2001) Of the Europa sur-face candidate materials that have been charac-terized at low temperatures thus far the hydratedsulfur compounds provide the best availablespectral matches (Carlson et al 1999b McCord etal 2001) to the hydrate features observed in theEuropa spectra Yet many materials have notbeen considered simply because they have not yetbeen measured (Dalton 2002b Jamieson and Dal-ton 2002) The most interesting candidates froman astrobiological perspective are those that arerelated to the occurrence of life Meteoritic andcometary delivery of carbon over geologic time ispredicted to eventually result in both inorganic(ie sodium carbonates carbon dioxide carbonmonoxide) and organic (ie amino acids form-aldehyde) carbon compounds through a com-bination of endogenic and exogenic processes

DETECTION OF MICROBES ON EUROPA 507

(Kvenvolden et al 1970 Kargel 1991 Delitskyand Lane 1997 1998 McCord et al 1999a Chybaand Phillips 2001 Kempe and Kazmierczak2002) Magnesium sulfate salts and sulfuric acid are both commonly found in terrestrial hotsprings environments which host extremophilicmicroorganisms (Rothschild and Mancinelli 2001and references therein) If the surface deposits onEuropa contain these compounds and these com-pounds are derived from an interior ocean it maybe that the deposits contain other materials fromthe interior If the interior once hosted life someremnant of this life may be entrained in the sur-face deposits If in fact the surface deposits didcontain remnants of biological organisms whatwould they look like today

Microorganisms possess unique infrared spec-tra because of their water of hydration and bio-molecular components Their spectra contain minor contributions from the absorption of func-tional groups (such as amides carboxylic acidsphosphates and alcohols) found in proteinslipids nucleic acids and carbohydrates and ma-jor absorption contributions arising from the wa-ters that hydrogen bond to and hydrate the bio-molecules We have measured the infraredspectra of four different biomolecules at standardtemperature and pressure and three terrestrialmicroorganisms at temperatures relevant to thesurface of Europa In this paper we compare thesespectra with those of disrupted terrains on Eu-ropa as measured by the Galileo NIMS instru-ment

Interpretation of these spacecraft observationsis critically dependent upon the spectral resolu-tion of the observations especially when exam-ining subtle spectral changes Because the mate-rials of interest are found in discrete surface unitsof disrupted terrain mixing of signal from adja-cent icy terrain can overwhelm the spectral in-fluence of the target material Thus high spatialresolution is just as important as high spectral res-olution In addition the signal-to-noise ratio de-termines the confidence levels that can be as-cribed to any spectral identification We haveevaluated the spectral correlations between labo-ratory and observational spectra in light of thesethree criteria The results of our investigation pro-vide valuable insights into the search for life onicy satellites and suggest methods whereby thenear-infrared biosignatures of life as we know itmay be used to constrain the astrobiological po-tential of these worlds

MATERIALS AND METHODS

This investigation was composed of three com-ponents (1) measurement and characterization ofinfrared spectra of microbes and biomolecules atvarying temperatures (measurements at 80ndash120 Kwere achieved using the specialized cryogenicvacuum equipment described below) (2) assess-ment of the survival of the microbes after expo-sure to the low pressures and temperatures and(3) comparison of the spectra with those of Eu-ropa as measured by the Galileo NIMS instru-ment

Infrared characterization of microbes and biomolecules

Sample selection Several model microorganismsand biomolecules were chosen for spectroscopicanalysis The microorganisms selected for this in-vestigation were Deinococcus radiodurans Sul-folobus shibatae and Escherichia coli D radioduranswas chosen because of its ability to survive inlow-pressure and high-radiation environmentssuch as those found in interplanetary space or atthe surface of planetary bodies D radiodurans cansurvive desiccation and ionizing radiation of 60Graysh for 30 h (1800 Grays 1 Gray 5 1 Jkg)with no loss of viability (Mattimore and Battista1996 Lange et al 1998 Fredrickson et al 2000Venkateswaran et al 2000) S shibatae was cho-sen as an analog to Europan microfauna becauseof its sulfur-based metabolism and the expectedhigh sulfur abundance at Europa (Kargel 19911998 Fanale et al 1999 Carlson et al 1999b2002) S shibatae thrives and can be cultured inacidic environments (Table 1) that contain highconcentrations of sulfur and sulfur compoundsincluding magnesium sulfate salts and sulfuricacid (Grogan 1989 Rothschild and Mancinelli2001) Taken together these attributes render themicroorganisms as ideal candidates for spectro-scopic study and comparison with Europa As acontrol E coli was also studied in order to drawdistinctions between the survival and spectralproperties unique to extremophiles

Spectra of the individual biomolecules that sig-nificantly contribute to the near-infrared absorp-tion of the microbes were measured at room tem-perature in order to provide insights into thespectral characteristics of the microbes Bovineserum albumin calf thymus DNA dextrin andL-a-phosphatidylcholine dipalmitoyl were cho-

DALTON ET AL508

sen to act as model protein nucleic acid carbo-hydrate and lipid samples respectively

Spectrometer system All laboratory spectra wereacquired using an Analytical Spectral Devices(ASD) portable field spectrometer This device cov-ers the visible and near-infrared spectral rangefrom 035 to 25 mm The visible and near-infraredportion of the spectral range (up to 1 mm) utilizesa silicon detector with a spectral sampling of 14nm and a resolution of 6 nm The remainder of thespectral range is covered by a pair of indium anti-monide (InSb) detectors with spectral sampling of2 nm and resolution of 11 nm (Goetz et al 1998)The ASD spectrometer was programmed to aver-age 100 measurements each having a 1ndashs integra-tion time together for each spectrum after check-ing for consistency five such spectra acquired ateach temperature were averaged to produce the re-sults presented here This corresponds to an inte-gration time of 500 s per observation

Cryogenic environment chambers Two differentcryogenic systems were utilized in this investi-

gation One system the Outer Solar System En-vironment Chamber (OSSEC) resides at the USGeological Survey in Denver CO the other is theNASA Ames Cryogenic Reflectance EnvironmentFacility (NACREF) in Mountain View CA Thoughthey differ in terms of cooling capability andmethods they are similar in other respects Bothuse sapphire (Al2O3) viewports to illuminate andto measure the sample both use stainless steelsample cups to minimize spectral contributionsfrom the equipment Both use silicon diode tem-perature probes for process monitoring and bothcan be sealed and operated either in vacuummode or with a dry nitrogen atmosphere for frag-ile samples Both are initially evacuated using ro-tary and sorption pumps but the OSSEC utilizesan ion pump to achieve high vacuum while theNACREF is outfitted with a turbomolecular pump

The OSSEC is cooled by means of a coldfingerattached to the sample holder which extends intoa liquid nitrogen bath directly below the sealedsample chamber The sample temperature is con-trolled by the addition or removal of liquid ni-trogen In practice the OSSEC reliably achievestemperatures as low as 80 K The NACREF fea-tures a closed-loop helium cryostat and com-pressor to cool the samples The temperaturediode readings are read by a programmable tem-perature controller which adjusts the tempera-ture by means of a resistive heater embedded inthe sample holder While temperature readingsfor both systems are good to within 1 K theNACREF provides reliable and convenient tem-perature control and can reach temperatures aslow as 20 K

The incidence and emission angles for both sys-tems are constrained by the viewport positionsThe OSSEC chamber is fixed at 5deg for both inci-dence and emission resulting in a phase angle of10deg The NACREF can be set for either normal in-cidence and emission or normal incidence and15deg emission and phase The latter was used forthe measurements in this paper The OSSEC is il-luminated by a 100 W tungsten halogen lampwith a Samlex regulated DC power supply TheNACREF uses a regulated DolanndashJenner fiber op-tic power supply and a custom tungsten-halogenlamp To prevent undesirable spectral effects ofdichroic reflector coatings commonly used in themanufacture of such lamps the reflector hous-ings in our lamps have been coated with a 275-mm aluminum film Light is collected at the sap-phire viewports for both chambers by a low-OH


TABLE 1 GROWTH MEDIUM FOR LABORATORY

CULTURES OF S SHIBATAE

Medium composition Amount

Sulfolobus medium (per 1 l)Yeast extract 2 gSucrose 2 g100 3 buffer 10 ml200 3 buffer 5 ml1000 3 buffer 1 mlH2SO4 02 mlDeionized H2O Remainder

100 3 buffer (per 1 l)(NH4)2SO4 130 gMgSO4 7H2O 25 gFeCl3 6H2O 2 g50 H2SO4 15 ml

200 3 buffer (per 1 l)50 H2SO4 5 mlKH2PO4 56 g1 MnCl2 36 ml1 ZnCl2 44 ml1 CuCl2 1 ml1 Na2MoO4 006 ml

1000 3 buffer (per 200 ml)CaCl2 2 H2O 14 g

The three buffer solutions are combined in the pre-scribed ratios with sucrose yeast extract and sulfuricacid then deionized water is added to make up 11 of solution having a pH of ~23 Note the inclusion of ammonium and magnesium sulfates and sulfuric acidcompounds predicted to occur on Europa

silica fiber optic which is standard to the ASDspectrometer

Sample preparation Powdered samples of thefour representative biomolecules were acquiredfrom reagent-grade stock and weighed under am-bient laboratory conditions Approximately 2 cm3

of each sample was placed in a steel sample cupand measured using the ASD spectrometer andfiber-optic illuminator at room temperature withSpectralonreg (Labsphere Inc) as the referencestandard

Cultures of D radiodurans (ATCC 51178) Sshibatae (ATCC 13939) and E coli strain W1485were grown at NASA Ames Research Center Dradiodurans was cultured in a TGY liquid medium(5 g of Tryptone 3 g of yeast extract and 1 g ofglucose per 1 l) and incubated at 30degC to mid-logphase S shibatae was aerobically grown in themedium described in Table 1 at 76degC up to latelog phase for about 4 days E coli was aerobicallygrown at 37degC in LB medium (1 Bacto-tryptone05 Bacto-yeast extract and 1 sodium chlo-ride) for 18 h Samples of each were centrifugedto produce 1 ml of densely packed cells andfrozen in a 280degC freezer

Microbial samples were transported on dry iceto the OSSEC facility in Denver Samples wereslowly thawed in a water ice bath and placed inthe OSSEC under a dry nitrogen atmosphere Thechamber was purged with nitrogen then sealedand rapidly cooled to 250 K and then evacuatedto 1024 Torr Spectra of each sample were ac-quired with the ASD at 20 K intervals from 200to 100 K Spectra of the S shibatae growth medium(Table 1) were also measured for comparison Thegrowth medium was virtually indistinguishablefrom pure water ice and therefore not a contrib-utor to the asymmetric absorption features con-sidered below After each series of spectral mea-surements the chamber was allowed to warm to250 K then pressurized to 1 atm with nitrogenStill frozen the samples were removed from thechamber transferred to sealed vials and placedon dry ice for the return to NASA Ames

Complications were encountered during spec-tral measurements because of migration of waterfrom between the cells to the tops of the samplesThe formation of water frost introduced spectralfeatures that obscured the spectral signal of thesamples Warming the sample up to 230ndash250 Kunder vacuum efficiently removed this waterhowever the D radiodurans sample was acciden-

tally dehydrated during this procedure As a re-sult the infrared absorption features arising fromwater of hydration in the D radiodurans were sig-nificantly reduced The measurement was laterrepeated using the NACREF and a new sampleof D radiodurans The same procedures were fol-lowed except that spectra were acquired every10 K and after spectra were acquired at 80 K thesample was cooled to 20 K Improved purging ofthe NACREF chamber with nitrogen and themore rapid cooling of the sample prevented frostmigration and eliminated the need to warm thesample after vacuum was established Upon re-moval from the chamber the sample was rapidlyfrozen in liquid nitrogen and placed in the 280degCfreezer to await survivability assessment

Low temperature survivability

After exposure to temperatures and pressurescorresponding to the polar latitudes of Europaall three samples were cultured to evaluate sur-vival rates Frozen cell pellets of D radioduransand E coli were resuspended in phosphate-buffered solution and the cell numbers were di-rectly counted under a light microscope using ahemocytometer D radiodurans cells were platedon TGY agar plates (TGY medium with 15 g ofagarl) and incubated at 30degC for 3 days E coliwas spread on MacConkeyrsquos medium plates andincubated at 37degC for 18 h Colonies appearing af-ter incubation were counted to calculate survivalrates Approximately 01 g of frozen S shibataecells were inoculated in 100 ml of medium andaerobically incubated at 76degC for 3 days After thecell number was counted the culture was dilutedat 1100 150 and 110 with fresh medium and in-cubated for 2 days Only the 110 dilution showedgrowth the cell number was counted to calculatethe survival rate Contamination issues were re-solved separately for each strain The S shibataeculture was grown in its acidic medium at hightemperature D radiodurans was identified by itsspecific reddish-pink colonies and E coli was se-lectively grown on MacConkeyrsquos plates whichcontain bile salts to inhibit the growth of most en-vironmental contaminants (Grogan 1989)

Survival rates were calculated separately forthe D radiodurans measured at the NACREF Re-sults from the first set of experiments indicatedthat most of the cells from all three samples hadbeen killed in the thawing and refreezing phasesnot in the subjection to Europa-like conditions

DALTON ET AL510

The second D radiodurans experiment did not in-volve long-distance transport or repeated freez-ing and thawing cycles The results of this secondexperiment indicated much higher survival rates

Processing of laboratory and spacecraft data

All room temperature and NACREF spectrawere acquired using Spectralon as the referencematerial while all measurements with the OSSECused Halonreg (Allied Chemical) as the referencethe influences of minor spectral variations in thereference materials were corrected using NationalBureau of Standards measurements for Spec-tralon and Halon Because of the long integrationtimes and high signal inherent in the laboratorymeasurements the error bars of the resultantspectra are too small to be clearly visible in thefigures and are therefore omitted for clarity

All laboratory spectra were processed usingthe SpecPr program (Clark 1993) and da Vincispectral math engine Vertical offsets at 10 and182 mm due to differences in gain states of theASD spectrometer detector electronics were re-moved by linear multiplicative scaling This pro-cedure preserves relative band shapes and depthsat the expense of absolute reflectance levels

Infrared spectra of Europa were obtained bythe Galileo NIMS instrument from 07 to 52 mmat a spectral resolution of 125ndash25 nm and sam-pling interval of 12 nm (Carlson et al 1992) TheNIMS observations (denoted G1ENNHILAT andE11ENCYCLOD) used in this paper are availableto the public and have not had any additional cor-rections or recalibrations beyond the standardprocedures applied by the Galileo NIMS TeamThese data were downloaded with 228 wave-length channels The G1ENHILAT observationwas acquired with nominal spatial resolution of77 kmpixel while E11ENCYCLOD was at 84kmpixel Twenty-six spectra of icy terrain and18 spectra of dark terrain in the G1ENNHILATdata were averaged together using SpecPr to im-prove signal-to-noise characteristics Averages of21 and 72 spectra from dark terrain and 160 spec-tra from icy terrain in the E11ENCYCLOD ob-servation were created to compare the spectralfeatures in the two terrain types It should benoted that the 72- and 160-spectra averages wereextracted from a version of the cube that had notbeen reprojected to remove spatial distortionsThis reprojection procedure may reduce spectralcontrast due to spatial averaging of different

spectral units This will be considered in the Dis-cussion Offsets between NIMS detectors were re-moved by the same linear multiplicative scalingapplied to the ASD spectra (Dalton 2000)

To facilitate comparison laboratory spectrawere convolved to the lower spectral resolutionand bandpass of the Galileo NIMS instrument using the Gaussian convolution routine in theSpecPr package Absorption band centers anddepths were calculated using standard contin-uum removal methods (Clark and Roush 1984)to eliminate interference from other absorptionfeatures in each spectrum and to remove depen-dence on the wavelength positions of the differ-ent spectrometer channels

RESULTS

Cryogenic spectra of microbe samples

The reasoning underlying this study of thecryogenic spectra of microorganisms involvedthe water of hydration bands in the Europa spec-tra While these spectral features are due to wa-ter this water must be in a bound state whichconstrains the permitted vibrational transitionsand gives rise to the distorted and asymmetricfeatures The difference between spectra of waterice and hydrated material is clearly demonstratedin Fig 1 where the unmistakable water ice fea-tures in the icy terrain are markedly different inthe spectra from the disrupted terrain The factthat these disrupted terrains are generally red-dish-brown is also of note While this reddishtone can be produced by iron and iron oxidesthere are strong absorption features near 08ndash1mm in the spectrum of iron which are not seenon Europa Complex sulfur polymers could giverise to similar coloration (Carlson et al 1999b2002) but the precise nature of these polymershas not yet been determined Hydrated salts can-not explain the coloration because they are white

An alternative explanation is the pigmentationthat is found in many microorganisms Microor-ganisms also contain significant amounts of wa-ter much of it as water of hydration Sulfur-metabolizing microbes such as S shibatae containsulfur compounds as well including sulfates Al-though the spectra of microbial samples mea-sured at room temperature contain water and wa-ter of hydration absorption features these are atthe wrong wavelength positions because theyarise from liquid water However at low tem-


peratures these shift to longer wavelengthsAgain the question becomes if microorganismswere emplaced at the surface of Europa whatwould they look like

The spectra of S shibatae D radiodurans and Ecoli at 120 K are displayed in Fig 2 The strongestabsorption features are those due to water of hy-dration near 10 125 15 and 20 mm These ex-hibit the characteristic asymmetry seen in thespectra of dark terrain on Europa Shortward of08 mm the spectra of all three drop off towardthe visible which is to be expected because oftheir reddish-brown coloration Finer structure isalso apparent in these spectra which are shownhere at the full resolution of the ASD instrumentNot all of this fine structure would be apparentat NIMS resolution A wide absorption is dis-cernible at 23 mm that is caused by the C-Hstretching mode common to spectra of many or-ganic compounds (Colthup et al 1990) The mostimportant differences from the hydrate featuresin the Europa spectra are the narrow features at205 and 217 mm These arise from amide bondswithin the peptide linkages of the cellular pro-teins (Fig 3) The 205-mm amide feature is a com-bination of the fundamental N-H vibration with

a C-N-H bending mode the 217-mm feature is acombination of the N-H fundamental with a C-Nstretching mode (Gaffey et al 1993)

The spectral characteristics of E coli are notmarkedly different from those of the extremo-philes Arising as they do from the same sorts ofbonds in the same sorts of compounds this is notsurprising This suggests that carbon-based mi-crobes will look quite similar in the near-infrareddespite major changes in functionality and har-diness While relative strengths and depths of individual absorption features may vary the features themselves remain the same It may beinferred that extraterrestrial microorganismscould bear strong resemblance to the ones pre-sented here

Infrared behavior of cell constituents

A better understanding of the sources of theseinfrared features can be gained from examiningspectra of individual cellular components Figure4 contains representative spectra of the four spec-trally dominant materials other than water itselfprotein nucleic acids carbohydrates and lipidsThese were all measured at room temperaturewith the ASD spectrometer All four materials areinfluenced by water particularly just short of 20mm but also near 15 mm The 125- and 10-mmwater absorptions are weaker and are not easilydistinguished in all of the spectra Some of thiswater is simply adsorbed on the powdered sam-ples The 15-mm water complex in the proteinand lipid spectra comprises at least three narrow

DALTON ET AL512

0

02

04

06

08

1

08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Escherichia coli

Deinococcus radiodurans

Sulfolobus shibatae

FIG 2 Near-infrared reflectance spectra of three mi-crobial samples at 120 K All three exhibit water absorp-tion features near 10 125 15 and 20 mm with the lat-ter two having the distorted and asymmetric naturecharacteristic of hydrates Additional features at 205 and217 mm are due to amide bonds in the cellular structuresSpectra have been offset vertically for clarity

N

H

CC

CO

FIG 3 Portion of a peptide chain This schematic di-agram illustates the molecular bonds that give rise to the205- and 217-mm absorption features in the microbespectra

combination bands (Pauling 1935) which tend tointeract more closely in pure water (and ice) tocreate the broad 15-mm feature seen in the icyEuropa spectrum of Fig 1 The diagnostic C-H

absorption at 23 mm is quite strong in the lipidspectrum and can also be identified in the pro-tein and carbohydrate spectra as expected

Spectral comparison of Europa candidate materials

The spectra of the primary Europa surface ma-terial candidates are shown in Fig 5 along withrepresentative spectra of S shibatae All of thesespectra were measured at 120 K and convolvedto the Galileo NIMS wavelengths and resolutioncorresponding to the G1ENNHILAT observation(bottom curve) Much of the structure evident inthe laboratory spectra is subdued at this resolu-tion The spectrum of sulfuric acid octahydrate istaken from Carlson et al (1999b) The spectra ofbloedite [Na2Mg(SO4)24H2O] and hexahydrite(MgSO46H2O) were taken from Dalton (2000)Based on room temperature measurements Mc-Cord et al (1998) suggested that bloedite andhexahydrite may together make up as much as65 of the surface composition in the disruptedterrains Carlson et al (1999b) proposed insteadthat sulfuric acid octahydrate is the dominant sur-face component All four materials exhibit dis-torted and asymmetric absorption features due towater of hydration near 10 125 15 and 20 mmIn the Europa spectrum the band centers are located at 101 123 148 and 195 mm (dottedlines) In hexahydrite all except the 20-mm fea-ture fall shortward of these positions Bloedite


0

02

04

06

08

1

12

08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Lipid

DNA

Protein

Carbohydrate

C-H

FIG 4 Room temperature spectra of the main infrared-active cellular components vertically offset forviewing Dotted lines mark the positions of the 205- and217-mm amide absorption features These are most preva-lent in the protein but are also visible in the DNA andlipid spectra The C-H stretch absorption feature near 23mm common to many organic molecules is apparent inthe lipid protein and carbohydrate spectra Water ab-sorption features are evident in all four spectra

FIG 5 Spectra of the S shibatae Ar-chaea sulfuric acid octahydrate and thehydrated sulfate salts bloedite andhexahydrite compared with the darkEuropa terrain spectrum from Fig 1Dotted lines denote positions of bandcenters from the Europa spectrum Oc-tahydrate spectrum is from Carlson et al(1999b) and was measured at 140 KBloedite and hexahydrite spectra (mea-sured at 120 K) are from Dalton (2000)The Galileo NIMS spectrum of Europais the same as in Fig 1 All spectra areconvolved to the NIMS wavelength setwith overlapping wavelength positionsdeleted for clarity

08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Europa

Hexahydrite

Bloedite

Sulfolobus Shibatae

H2SO4 8H2O

provides a better match for the first three bandsbut the 20-mm feature is centered at a slightlylonger wavelength Sulfuric acid octahydratematches well at 125 and 15 mm (no data wereavailable below 10 mm) but the 20-mm absorp-tion occurs at a longer wavelength as in the saltsSome of these wavelength discrepancies andchanges in band shape may be attributable to ra-diation damage (Nash and Fanale 1977) reduc-ing the importance of exact matches Still it is in-teresting to note that the first three absorptionband centers of S shibatae fall slightly longwardof the Europa positions but with smaller dis-placements than are observed in the hydratedsalts The 20-mm band center in the S shibataespectrum matches the Europa position exactly

Each measured spectrum also displays featuresnot observed in the Europa spectrum Hexahy-drite and bloedite both exhibit additional bandswithin the 15-mm complex some of which aredue to separation of combination bands at lowtemperatures Hexahydrite has a weak band at22 mm while bloedite has a stronger one at 21mm Sulfuric acid octahydrate does not exhibitany additional strong absorption features thoughit and the salts all have a cation-OH stretchingfeature at 135 mm which is discernible at NIMSresolution (Dalton and Clark 1999) Three fea-tures not apparent in the Europa spectrum of Fig5 are displayed by all three microbial samples asrepresented by S shibatae the amide absorptionfeatures at 205 and 217 mm and the C-H stretchat 23 mm

Because the relative depths widths shapesand positions of the 10- 125- and 15-mm fea-tures in the Europa spectrum can all be repro-duced individually using only water ice of vary-ing grain sizes (McCord et al 1999a Dalton2000) the 20-mm absorption feature was used asa diagnostic signal To highlight this informationthe standard continuum removal method ofClark and Roush (1984) was applied to each ofthe laboratory spectra and to the observed spec-trum from Fig 5 This reduces interference fromcontinuum absorption and transitions outside thewavelength range of interest The resulting con-tinuum-removed spectra are plotted in Fig 6with vertical offsets applied for visibility The cor-respondence between the band shapes widthsdepths and centers in the laboratory and obser-vational spectra can be ascertained more readilythis way Viewed in isolation the 20-mm ab-sorption band complex reveals a number of di-

agnostic characteristics The band center in theEuropa spectrum at 195 mm (dotted line a) ismost closely matched by the S shibatae spectrumwhile the three hydrate spectra are slightly longerin wavelength indicating a lower vibrational fre-quency than that observed at Europa The addi-tional absorption features at 21 mm in bloediteand 22 mm in hexahydrite are not evident in theEuropa spectrum though a decrease at 22 mm inthe Europa spectrum may be significant How-ever this slight decrease is only two channelswide and may therefore be due to noise Thehexahydrite feature is much wider six channelsat the resolution of the NIMS observation

More telling perhaps would be evidence of theamide features at 205 and 217 mm (dotted lines

DALTON ET AL514

18 19 2 21 22 23

05

1

15

Co

nti

nu

um

-co

rrec

ted

ref

lect

ance

Wavelength (mm)

Europa

Hexahydrite

Bloedite

H2SO4 8H2O

Sulfolobus Shibatae

a b c

FIG 6 Continuum-corrected spectra of the 20-mm ab-sorption feature for the materials and wavelengths ofFig 5 Dotted lines denote center positions of (from leftto right) (a) 20-mm water of hydration feature in Europaspectrum and (b) 205- and (c) 217-mm amide features inS shibatae spectrum (arrows) Error bars denote standarddeviation of the mean for the original 18ndashspectrum aver-age (Dalton 2000)

b and c) The feature at 205 mm appears to cor-respond to a similar feature in the Europa spec-trum Though close to the noise level of the ob-servations this 3 band depth feature (see Table2) is 80 nm (six channels) wide A slight inflec-tion at 217 mm is barely discernible in the NIMSobservation This absorption feature has a banddepth of 5 and a width of 120 nm (nine chan-nels) which should make it easier to detect Un-fortunately the short wavelength edge of this fea-ture overlaps with a gap between detectors in theNIMS instrument At such positions (note theclosely spaced data points near 184 and 212 mm)the absolute reflectance cannot be determinedprecisely because the NIMS detectors switch fromone grating step to the next introducing a gainshift that is difficult to accurately compensate(Carlson et al 1992 McCord et al 1999b) Thisrenders any identification suspect A more robustidentification could be performed with higherspectral resolution and signal levels Despitethese uncertainties the wavelength positions ofthe apparent features and of the 20-mm featureitself align remarkably well with those of the Sshibatae measured at 120 K Within the limits ofnoise and spectral resolution discussed belowthe shape of the 20-mm band complex in Fig 6comes close to what one might expect to see if mi-croorganisms were in fact trapped near the sur-face in the disrupted reddish terrain of Europa

Search for infrared biosignatures on Europa

To date examination of the Galileo NIMS andSSI data has produced few examples of thisbiosignature (Dalton 2002a) The modest spectraland spatial resolution of the NIMS data makes itdifficult to locate spectra derived primarily fromstrongly hydrated material and exhibiting suffi-cient signal to identify these subtle absorptionsBecause the wavelengths are not identical in allNIMS observations the NIMS channels are notalways ideally situated for identifying narrowfeatures

One observation that may represent potentialevidence for biosignatures is the E11ENCY-CLOD01 image cube described earlier A false-color image in Fig 7 depicts the NIMS observa-tion overlaid on an SSI mosaic The SSI image has a spatial resolution ranging from 300 to 220mpixel The much larger NIMS pixels (84 kmnominal) have not been smoothed or averaged inany way While the instantaneous spatial profileis not square the effective field of view is (R Carl-son personal communication 2003) so these rawpixels give a reasonable indication of the locationon the surface being sampled Each pixel repre-sents a single spectrum (ldquoNIMSELrdquo) correspond-ing to that location In the false-color image blueindicates icy spectral behavior in the 20-mm re-gion while red corresponds to the highly asym-


TABLE 2 ABSORPTION BAND CENTER POSITIONS DEPTHS AND WIDTHS FOR THE THREE MICROBES MEASURED AT 120 KAND FOR THE TWO EXTRACTED SPECTRA FROM DARK TERRAIN IN NIMS OBSERVATION E11ENCYCLOD01

Absorption feature center position depth and width

20-mm H2O 205-mm amide 217-mm amide

D radiodurans 19468 20525 21718 Center (mm)02602 00284 00515 Depth01593 00310 00531 Width (mm)

S shibatae 19577 20603 21694 Center (mm)03853 00688 00258 Depth02082 00455 00463 Width (mm)

E coli 19556 20535 21744 Center (mm)01241 00141 00176 Depth01734 00343 00495 Width (mm)

72 NIMSEL 19582 20459 21814 Center (mm)04983 00264 00238 Depth01954 00201 00130 Width (mm)


All values were calculated using the methods of Clark and Roush (1984) Band centers are for continuum-correctedspectra The band depth is defined as the ratio of the reflectance at the band center position divided by the reflectanceof the fitted continuum at the same wavelength Widths are FWHM by the same method

metric absorption features associated with hy-drated material This hydrated material isstrongly correlated with the linea and some of thedark spots in the image Few areas are composedof pure water ice Spectra were extracted fromvarious regions within the image cube and ex-amined for evidence of 205-mm features Wherepossible spectra from contiguous regions wereaveraged together to improve signal-to-noise ra-tios This was complicated by the high degree ofspatial heterogeneity at small scales on EuropaEven at 220-m resolution features can be dis-cerned in the SSI image that are smaller or nar-rower than the SSI pixels so finding homoge-neous areas the size of the NIMSELS provedchallenging For example the central stripes oflinea tend to be composed of icier material thanthe margins so linea spectra typically have an icycomponent to them In the end many of the spec-tral averages were partly contaminated by waterice due to subpixel-scale mixing

One area of 21 contiguous NIMSELS (greenFig 7) proved interesting The averaged spectrumis shown in Fig 8 (green curve) This spectrum isdominated by the asymmetric 20-mm hydratefeature An apparent feature at 205 mm takes upseveral channels and there is a slight indentationat 217 mm as well Because the reprojectionprocess involves some spectral averaging it mayreduce both spectral contrast and variations insignal resulting in weaker features but smallernoise levels It may also introduce spurious fea-tures that typically arise from a single-channelnoise spike being averaged into neighboringspectra Spectral contrast may be reduced whena spectrum having a specific absorption featureis averaged with a spectrum in which that featureis absent An average of 21 spectra is less likelyto have spurious features but given the noise lev-els of the NIMS spectra such an average is stillsuspect

The unprojected image cube is shown at thetop of Fig 7 Spacecraft motion and viewinggeometries produce considerable spatial ldquosmearrdquoin the observation but the spectra are more rep-resentative of surface material than those fromthe reprojected cube Major surface features in-cluding the intersecting linea can still be distin-guished in the unprojected cube Because there isnot a one-to-one correspondence between pixelsin the projected and unprojected cubes no set ofNIMSELS in the unprojected cube was able to re-produce the spectrum just described Where the

desired spectral features were present noise inother parts of the spectrum made interpretationdifficult However examining other parts of thecube revealed similar features in the other lineaThe southeastern linea is covered by more NIMSELS and in particular is more than oneNIMSEL wide along much of its length This al-lows a larger number of samples of nonndashwater iceterrain As demonstrated in McCord et al (1999b)the capabilities of the NIMS instrument com-bined with the spatial heterogeneity and high ra-diation environment at Europa necessitate spec-tral averaging of carefully inspected NIMSELS inorder to produce spectra that accurately reflectthe surface constituents of interest Followingtheir method 72 NIMSELS (indicated in red Fig7) were selected and averaged in Fig 8 SomeNIMSELS contained noise spikes within thewavelength range of interest and were discarded(black pixels Fig 7) In order to compare the spec-tra of the heavily hydrated linea material with themore icy background plains a separate averagewas made using 160 NIMSELS (blue pixels) lo-cated between the major linea While some dis-rupted areas and narrow linea traverse theseplains NIMSELS showing strong features of hy-dration were omitted along with those havingstrong noise spikes (black pixels)

No offset smoothing or scaling has been ap-plied to the spectral averages in Fig 8 Overlap-ping wavelengths at the spectrometer gratingsteps have been retained All error bars indicate1-s levels of standard deviation of the mean Thedifferent amplitudes of error estimates for the 21-versus the 72- and 160-NIMSEL averages reflectadditional processing of the data in the reprojec-tion method The 205-mm absorption feature is somewhat weaker in the 72-NIMSEL averagethan in the 21-NIMSEL average (25 of re-flectance as opposed to 3 Table 2) and does notappear to be quite as broad Of course this is adifference of only two channels but when deal-ing with this few channels it affects the conclu-sions that can be drawn Similarly the 217-mmabsorption feature if present can only occupy afew channels The 72- and 21-NIMSEL averagesboth have an apparent feature at this location ex-cept that one channel is high (219 mm) com-pared with its neighbors This would seem to ar-gue against an identification based on thisfeature If this single channel were high becauseof random noise then neighboring channelsshould not be affected On the other hand the ap-

DALTON ET AL516


210 W

20 N

10 N

NIMS E11ENCYCLOD

FIG 7 Galileo NIMS E11ENCYCLOD01 observationTop Unprojected NIMS cube with colored pixels indi-cating NIMSELS extracted for spectral averages in Fig 8Blue 160-NIMSEL average of icy plains units red 72NIMSELS of dark linea material black NIMSELS havinglarge noise spikes in the 20-mm range which have beendropped from the average Bottom NIMS E11ENCY-CLOD observation in Mercator projection overlaid on theGalileo SSI image to illustrate locations of dark disruptedsurface material Twenty-one NIMSELS extracted fromdark linea terrain in the projected image have been aver-aged in Fig 8

0

005

01

015

02

18 19 2 21 22 23

NIMS E11 ENCYCLOD01 Observation

Ref

lect

ance

Wavelength (mm)

72 NIMSEL AVERAGE

21 NIMSEL

160 NIMSELICY REGION

FIG 8 Averaged spectra from observation E11ENCY-CLOD01 showing potential evidence of biological ma-terial in disrupted terrainA feature corresponding to the205-mm amide absorption is apparent in the two spectrafrom the dark linea in Fig 7 (red and green curves) Theblue curve an average of 160 spectra from the area be-tween the linea has a more symmetric 20-mm absorptionband and no evidence of features at 205 and 217 mm Afew channels are low at 217 mm in the dark terrain curvesbut one channel is high (at 219 mm) making confident as-signment of this feature difficult

pearance of the same spectral behavior over sev-eral channels in both averages might indicate asystematic effect possibly due to instrumentcharacteristics Yet both averages are from dif-ferent locations in the observation and the 160-NIMSEL average shows no such feature In factthe 160-NIMSEL average does not exhibit evi-dence of either feature which suggests that thesefeatures may be linked to geographic units afterall Although it was collected over primarily icyterrain the 160-NIMSEL average does exhibitsome asymmetry in the 20-mm absorption fea-ture and therefore some hydrated material is pre-sent However the 205- and 217-mm absorptionsare not apparent in the 160-NIMSEL average con-sistent with a predominantly icy surface Exami-nation of spectra throughout the E11ENCYC-LOD01 E6ENSUCOMP02 and E4ENSUCOMP03NIMS observations did not identify evidence ofthese features although regions of highly hy-drated material were studied This indicates thatthe 205- and 217-mm absorptions are not foundin all of the dark disrupted hydrated terrainThis is not surprising since the material givingrise to this spectral signature may not be homo-geneously distributed Just as the dark materialappears darker and more concentrated in someareas there may be processes acting to concen-trate other constituents within the dark terrains

Survival of organisms subjected to extreme cold

The ability of microorganisms to survive ex-posure to extreme conditions has implications notjust for Europa but for the search for life in theUniverse itself particularly as it relates to the hy-pothesis of panspermia Survival and even meta-bolic activity have been quantified at subzerotemperatures and certain cells have been foundto survive temperatures approaching absolutezero (Mazur 1980 Rivkina et al 2000) Indeedstorage of frozen and freeze-dried microorgan-isms is a standard practice in microbiology re-search (Hay 1978 Freshney 2000)

Culturing of our samples after exposure to Eu-ropan temperatures during cryogenic spectro-scopic measurements revealed viable microor-ganisms in each sample as determined by thelimiting dilution method described in the previ-ous section The S shibatae proved to be the mostfragile with a survival of 1 cell out of a culturecontaining between 46 3 109 and 92 3 108 total

cells (22 3 10210 and 11 3 1029 living cellsto-tal cells) The hardiest of the three was D radio-durans with a survival of 5 3 1022 living cellstotal cells E coli although not considered an extremophile had a survival rate of 34 3 1028

living cellstotal cells one order of magnitudehigher than that of S shibatae Control samples ofD radiodurans and E coli which had been iden-tically stored but not spectrally analyzed werealso measured for survivability Respective sur-vival rates for these samples were 5 3 1022 and47 3 1028 living cellstotal cells This indicatedthat cell death occurred primarily because of thefreezing and thawing that occurred during sam-ple preparation and not because of the extremelycold temperatures of the experiment

Survival of D radiodurans analyzed by theNACREF was also measured at temperatures rel-evant not only to Europa (120 K) but also inter-stellar space (20 K) Two aliquots were removedafter exposure to the extreme conditions cul-tured and measured for survivability An aver-age survival of 71 was achieved after exposureto temperatures and pressures similar to those en-countered in interstellar space Because D radio-durans is known for its robust nature this maynot be surprising The two major points to bemade here are (1) that survival rates were non-negligible for even the extreme cold of interstel-lar space and (2) that even E coli which is notparticularly well adapted to such harsh condi-tions exhibited some resistance to destruction byrepeated thawing and freezing to extreme tem-peratures These results have important implica-tions for the survival of microorganisms at Eu-ropa and in space

DISCUSSION

The application of cryogenic spectra to theanalysis of remotely sensed spectra of icy satel-lites provides unique insights into planetaryprocesses For this study the comparison ofGalileo NIMS spectra with the spectra of ex-tremophilic microorganisms measured in the lab-oratory has implications for the interpretation ofthe Galileo data Several effects must be takeninto account when interpreting these results in-cluding the intense radiation environment and itseffects on the survival of not just microorganismsbut of any organic matter Additional work willbe needed to resolve some of these issues

DALTON ET AL518

Comparison of Galileo and laboratory spectra

The low-temperature spectra of microbial sam-ples in Figs 2 and 5 display all of the pertinentfeatures of hydrates The primary water featuresnear 10 125 15 and 20 mm are extremely sim-ilar to those seen in the NIMS Europa spectra Thisin itself is not sufficient to claim an identificationsince other materials may also display similarspectral effects The hydrated materials in Fig 5also have hydrate features at these positions Theoverall slopes of the spectra tend to match fairlywell although the slope for the sulfuric acid oc-tahydrate is somewhat too steep The band centerpositions for the microbe specimens as repre-sented by S shibatae in Fig 5 are also much bet-ter matches to the Europa spectrum This is sig-nificant because the wavenumber of an absorptionis directly proportional to the energy differencebetween molecular states and the energies ofthese states is proportional to the vibrational fre-quency of the molecule If the vibrational fre-quency does not match the observational evi-dence then the molecular bond lengths bondstrengths and component atoms cannot be iden-tical It may be possible to reproduce some of theseobserved band shifts in the laboratory throughcharged particle irradiation effects which alter thechemical and physical properties of the candidatematerials While the relative depths of manybands are closer to the Europa proportions in themicrobe spectrum this is not as important as bandcenter position because relative depths can bestrongly influenced by particle grain size effectsscattering from grain boundaries and mixtureproportions These effects can be difficult to du-plicate in the laboratory Much more telling is thereduction in reflected intensity shortward of 1 mmin the Europa and S shibatae spectra This drop offis not present in the hydrated salts (Hunt et al1971 Crowley 1991) but is evident in the spectraof all three microorganism samples (Fig 2) Oneadditional consideration is the visible reflectanceD radiodurans E coli and S shibataewere selectedfor this experiment partly because of their red-dish-brown coloration which is similar to thatseen on Europa All of the postulated hydratedsalts and acids are white Even so this does noteliminate them from consideration as it may yetbe possible to reproduce the observed colorationthrough irradiation of hydrated sulfur com-pounds Further work will be necessary to inves-tigate this possibility

Examination of the continuum-removed spec-tra in Figs 9 and 10 clarifies the types of spectralbehavior that would characterize the biosigna-tures of entrained microorganisms at the surfaceof an icy satellite The asymmetric 20-mm ab-sorption feature dominates the spectra of all threemicroorganisms and the two spectra of reddishdisrupted terrain on Europa There is also someasymmetry in the 160-NIMSEL average of spec-tra from the ldquoicyrdquo plains in the NIMS observa-tion which indicates the presence of some hy-drated material in spectra making up the average(Fig 7) The 205- and 217-mm amide features inthe microbe spectra (Fig 9) vary in their relativeintensities Again this can be because of scatter-ing from water crystals changes in ice grain sizesor differences in cell composition andor con-stituent concentrations Each of these is difficultto control Note that while the 217-mm feature ismuch weaker than the 205-mm feature in the Sshibatae spectrum the opposite is the case in theD radiodurans spectrum Both bands appear tohave similar strengths in the E coli spectrumWhile the depth of the 217-mm feature rangesfrom 17 to 52 of the continuum level withinthe 20-mm complex the 205-mm feature ranges


05

06

07

08

09

1

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)

E coliD radioduransS shibatae

FIG 9 Continuum-corrected spectra of E coli D ra-diodurans and S shibatae measured at 120 K The 20-mm water of hydration band center lies closer to 195 mmThe amide absorption features at 205 and 217 mm varyin strength among the three samples No offset has beenapplied after the continuum removal

between 14 and 69 Both are weakest in theE coli spectrum The relative depths of thesebands with respect to each other can thus varysignificantly in terrestrial microorganisms andcan therefore be expected to do so in extraterres-trial microorganisms as well More work shouldbe done to quantify these variations and theircauses At present it would appear that the oc-currence of these bands in concert with the 20-mm hydrate feature could be construed as a pos-sible biosignature

The shapes of the continuum-corrected NIMSspectra in Fig 10 are suggestive of a 205-mm fea-ture and possibly a 217-mm feature As these fea-tures are superposed upon the 20-mm band com-plex the continuum-removed spectra clarify theirrelative contribution There is clearly no evidencefor either feature in the icy terrain spectrum eventhough it shows the influence of hydrates Theband depth of the 205-mm feature in the 21-NIM-SEL average is 63 while only 26 in the 72-NIMSEL average (Table 2) The weaker signal inthe 72-NIMSEL average could be due to compo-nent spectra that do not exhibit the feature (iehydrated material that does not contain amides)The evidence for a 217-mm feature is evenweaker particularly with the high channel at 219

mm The apparent depth of this feature is 2though slightly higher (24) in the 72-NIMSELspectrum If the 219-mm channel is anomalouslyhigh this depth could be much greater but suchestimates are highly speculative

The high radiation environment at Europamakes interpretation of narrow spectral featuresparticularly challenging A single radiation hit ona detector may manifest itself as a single channelspike often as far as five times the calculated vari-ance of the neighboring channels but frequentlymuch smaller and occasionally well within thevariance For this reason spectral averaging ofneighboring pixels has been used (McCord et al1998 1999b Dalton 2000) to boost signal levelsAs described earlier this can be troublesome fortwo reasons First neighboring pixels may not berepresentative of the material in the original pixelthe high degree of spatial heterogeneity at smallscales on Europa becomes critical Second spec-tral averaging can alter feature strengths A noisespike may propagate into spectral averages andcreate apparent features that are not there or al-ternatively a feature may be averaged away bythe influence of pixels where it is either not pre-sent or is obscured by random noise

Although some pixels with obvious noisespikes outside the 3s level were omitted from thespectral averages (black pixels in Fig 7) some se-lection effects may remain The spectral averagefrom the projected E11 NIMS cube may sufferfrom errors introduced in the projection processArtifacts of the instrument and instrument cali-bration may obscure the spectral signal The largenumber of pixels in the unprojected averages (72and 160) combined with the presence of the ap-parent features in both 72- and 21-NIMSEL aver-ages would argue against purely random noiseLikewise the similar shapes of the apparent 217-mm feature in both averages yet its absence fromthe 160-NIMSEL ldquoicyrdquo average would seem to ar-gue against a systematic error

Given the low apparent strength of these fea-tures and the high noise levels we cannot confi-dently interpret them as evidence of life on Eu-ropa In particular the fact that these features areso narrow coupled with their placement withinthe much stronger hydrate absorption featuremakes their detection and verification challeng-ing at the sampling and bandpass of the NIMSEuropa observations The 205-mm amide featurehas a full width at half-maximum (FWHM) of be-tween 30 and 45 nm in the microbial samples

DALTON ET AL520

0

02

04

06

08

1

12

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)

72 NIMSEL21 NIMSEL160 NIMSEL

NIMS E11 ENCYCLOD01

FIG 10 Continuum-corrected spectra from NIMSE11ENCYCLOD01 Each has been vertically offset for vi-sual clarity The absorption features at 20 205 and 217mm correlate well with those in the microbe spectrashown in Fig 9

while the 217-mm feature has a FWHM from 46to 53 nm To reliably detect these features evenunder low noise conditions would require sam-pling of at least 10 nm which would place onlyfive channels across their FWHM Given the highradiation environment at Europa 2ndash5 nm sam-pling is needed to significantly increase the po-tential for robust discrimination of these biosig-natures against the background continuum

Radiation survival of biological material

The radiation environment at Europa alsoposes specific challenges not just to the detectionof life but to the survival of the very compoundsthat make up most biological matter Bombard-ment of the surface by energetic particles en-trained within the Jovian magnetosphere drivesa number of physical and chemical processes thatare detrimental to the persistence of complexmolecules Ironically the radiolysis driven byparticle bombardment has also been invoked asa method to produce compounds that could beused as energy sources by potential Europan lifeforms (Chyba and Phillips 2001 Borucki et al2002) The heavy-ion and proton bombardmentflux is calculated to be between 108 and 1010 par-ticles(cm2 s) (Ip et al 1998 Cooper et al 2001)The bulk of the energetic deposition occurs in theform of electrons with energies in the 2ndash700 keVrange with a mean energy near 350 keV (Cooperet al 2001) These electrons interact with the sur-face water ice to produce H and OH radicals aswell as highly reactive species including O O2and HO2 (Hochanadel 1960 Johnson and Quick-enden 1997 Carlson et al 2002) Other elementsdelivered via the magnetospheric particle flux in-clude presumably Iogenic sulfur sodium potas-sium and chlorine (Carlson et al 1999b 2002Cooper et al 2001 Paranicas et al 2001) Com-plex interactions among the incident particlestheir secondary radiolysis products and the sur-face produce a number of derivative products(Moore 1984 Delitsky and Lane 1997 1998 Carl-son et al 1999ab 2002)

Organic matter andor microorganisms ex-posed to this particle flux are expected to be bro-ken down over geologic timescales Earlier it wasmentioned that D radiodurans can survive expo-sures of up to 1800 Grays this corresponds toless than a 3-day exposure at 1 mm depth on Eu-roparsquos surface based on dose rates from Parani-cas et al (2002) Recent experiments using low-

temperature plasmas have demonstrated the abil-ity of oxygen plasmas to completely degrade Dradiodurans and biomolecules into CO2 N2 andH2O (Bolrsquoshakov et al 2003 Mogul et al 2003)However these experiments also showed that Dradiodurans was not significantly degraded by aplasma of a Martian atmospheric gas analogThese experiments highlight the importance ofplasma composition on the reaction rates of mi-crobial degradation The surface gas compositionand ion densities at Europa would suggest thatthe rate of destruction is not extremely fast Fur-thermore since much of the destruction at thesurface of Europa is due to radiation-inducedchemistry and not collisional in nature the re-duced reaction rates prevailing at cryogenic tem-peratures cannot be neglected

In the absence of detailed studies concerningradiolysis of biological material at cryogenic tem-peratures some preliminary calculations may bemade based on other available information Mc-Cord et al (2001) have suggested that the resis-tance of various compounds to destruction by radiolysis may be estimated based upon thestrengths of the various molecular bonds withinthe compounds They reason that because the hy-drogen bonds associated with the waters of hy-dration in many of the hydrated salts are of equalor greater strength than those in water ice (Eisen-berg and Kauzman 1969 Novak 1974) the hy-drated salts should be at least as stable as waterice at the surface of Europa To support this ar-gument they also point out that the energies ofthe water infrared absorption features are simi-lar Following this line of reasoning one may con-sider the strengths of the amide bonds that giverise to the 205- and 217-mm absorption featuresFirst for comparison the 20-mm water feature it-self is a combination of the n2 H-O-H symmetricbending and the n3 H-O-H asymmetric stretchingmodes These involve the O-H bonds which havebond energies of 494 kJmol (Table 3) The 205-mm amide feature is also a combination bandcomprising the nN-H fundamental in concert witha C-N-H bending mode (see Fig 3) while the217-mm amide feature combines the nN-H funda-mental with a C-N stretching mode (Colthup etal 1990 Gaffey et al 1993) The bonds within thepeptide linkage of Fig 3 and their energies aregiven in Table 3 Examination of Table 3 illus-trates that the bonds responsible for the amidefeatures are all at least two orders of magnitudestronger than the hydrogen bonds of water ice


and hydrates and closer in strength to the cova-lent O-H bonds of the water molecule This sug-gests the possibility that at low temperature theamide bonds may be resistant to destruction byradiation-induced chemistry enough so to sur-vive to the current epoch and possibly even moreso than the hydrated salts as discussed in McCordet al (2001)

The suggestion that radiation resistance of acompound can be estimated from molecular bondstrengths rests upon a number of assumptionsand does not take into account the fact that radi-olytic reactions involve electronic excitations ofmany electron-volts which can induce dissocia-tion from excited states A deeper understandingof stability issues can be gained from examina-tion of the underlying mechanisms The radi-olytic destruction of most solutes in dilute aque-ous solutions occurs primarily through thediffusion and attack of hydroxyl radicals and sol-vated electrons which are produced by the radi-olysis of water This is due to the fact that the solvent (water) is in a much higher concentra-tion than the solute of interest (Garrison 1968Kminek and Bada 2001) The cold temperaturesand non-liquid state of the Europan ice crustwould limit the diffusion of chemically reactivespecies produced through radiolysis and there-fore increase the apparent lifetime of the biomarkerRadiolysis studies on formamide (HCONH2) havedemonstrated that G values (yield per 100 eV ofdeposited energy) for all products are 1 andthat the amide bond is actually retained in someof these radiolysis products (Matsumoto et al1970) Investigations of peptides and proteinsshow that radiolysis of peptides also results in re-tention of the amide bond along with formationof NH3 CO2 H2O2 keto acids and aldehydes(Garrison 1987) The respective G values for these

products are dependent on temperature radia-tion dose O2 concentration and the availabilityof radical scavengers Several of these variablesare not yet constrained well enough to calculateamide lifetimes at Europa based on existing lab-oratory studies Qualitatively however it is ap-parent from chemical models and laboratorystudies that the lifetime of the amide bond maybe much higher than those of the C-H and hy-dration bonds particularly at low temperatures(Matsumoto et al 1970) The approximate life-times and G values need to be measured underconditions that duplicate the temperature den-sity and radiation dose at Europa and will be thesubject of future studies The outcome of suchstudies may address a fascinating considerationIf the detritus of radiation-processed microbes re-tains both hydrate and amide features then thiscould be a powerful biomarker to be utilized inthe search for life on Europa and other icy satel-lites If however the amide bonds are easily de-stroyed by radiation then the resulting detritusmay provide an even better spectral match to theGalileo observations than any other candidatematerial considered thus far

Whatever their ultimate nature radiolysisproducts and implanted species are gradually in-corporated into the surface Cratering rates andestimates of impact gardening suggest that theuppermost 13 m of the surface can be overturnedand well mixed on a geological resurfacingtimescale of approximately 10 Ma (Zahnle et al1998 Chyba 2000 Cooper et al 2001) Biogenicelements delivered by comet impacts may havebeen mixed all the way down into the ocean overthe age of the Solar System providing enoughraw material to produce microorganism densitiesas high as 5 3 103 cellscm3 (Whitman et al 1998Pierazzo and Chyba 2002) This calculation as-

DALTON ET AL522

TABLE 3 BOND ENERGIES FOR MOLECULES CONSIDERED AS EUROPA SURFACE CANDIDATES IN THIS PAPER

Molecule Bond Bond energy (kJmol)

H2O H O 4937 6 08Ices hydrates Hydrogen 20Acid salts [ie MH(RCOO)] Hydrogen 25ndash35Peptides proteins amino acids C O 5259 6 04

C C 498 6 21C H 335 6 4C N 2782N H 372 6 8

Sources Cottrell (1958) Pauling (1960) March (1968) Eisenberg and Kauzman (1969) Darwent (1970) Novak (1974)and Joesten and Schaad (1974)

sumes a global average and neglects mechanismsfor concentrating either raw materials or livingsystems Convective overturn circulation pat-terns diapirism rifting and lithospheric brinemobilization caused by subsurface heating are allmechanisms that may contribute to the concen-tration of biogenic elements or even biologicalmaterial both within the icy crust and in the in-terior ocean (Spaun et al 1998 Greenberg 1999Head and Pappalardo 1999 McKinnon 1999McKinnon and Shock 2002 Prockter et al 2002)It is likely that some process would be necessaryto concentrate life-sustaining materials in orderto support life within the ocean determining thenature and viability of such mechanisms is be-yond the scope of this paper

There may be other substances present in thesurface that could account for the observed spec-tral signature As mentioned earlier there are aplethora of hydrated compounds whose spectrahave not been measured in the laboratory forcomparison with the icy satellites This is a con-sequence of the relative infancy of the field ofplanetary remote sensing and of the difficultyinherent in reproducing the relevant temperatureand pressure conditions Given the radiation en-vironment at Europa it is reasonable to ask whatsorts of compounds might be expected at the sur-face as a result of chemistry involving either ex-ogenically derived or endogenic species Irradia-tion of sulfur compounds may play a role inproducing some of the observed spectral shiftsand polymeric sulfur rings and chains (S8 Sx)have been suggested as possible coloring agents(Nash and Fanale 1977 Carlson et al 1999b2002) The 205- and 217-mm amide bonds do notof necessity imply viable microorganisms at thesurface These signals may also arise from for-mamide or simple dipeptides Pierazzo andChyba (1999 2002) have considered the potentialfor amino acids to be delivered to Europa bycomets and concluded that impact survival ofsuch complex molecules is likely to be negligibleThe apparent correlation however of the amideabsorption features with disrupted terrain is ofinterest since it indicates that while radiation pro-cessing may play a role there is certainly endo-genic material associated with the observed sur-face features There is also the possibility of amineral or similar substance that could explainthe observed spectral features Some OH-bearingminerals such as phyllosilicates and various hy-droxides exhibit spectral absorptions near 22 mm

(Hunt et al 1973) which could explain the spec-tral behavior in the NIMS observations shown inFigs 6 8 and 10 The apparent structure at 218and 221 mm in the NIMS spectra could be due toa number of materials that could obscure poten-tial evidence of amide absorptions in this rangeAlunite [KAl3(SO4)2(OH)6] for example has acombination feature at 217 mm caused by the fun-damental O-H stretch together with an Al-O-Hbend (Hunt et al 1971) Several alteration miner-als formed in the presence of sulfur compoundsunder terrestrial conditions are known to haveabsorption features between 216 and 223 mm arising from this combination Similarly Mg-O-H-bearing minerals sometimes exhibit fea-tures at 21 22 andor 23 mm (Hunt and Ash-ley 1979 Gaffey et al 1993) It is possible thatsome other oxide or hydroxides on Europa couldgive rise to the 205- and 217-mm features Dam-age from irradiation may further complicate thepicture inducing wavelength shifts in absorptionpositions Whatever alternative compounds maycome under consideration it must be remem-bered that to explain the observations the surfacematerials must exhibit both amide features aswell as water of hydration

Implications of survival

The survival of cultures exposed to low tem-peratures and the survival of biological materialexposed to particle bombardment both have far-reaching significance Microbiologists may not besurprised to read of the results of our survival ex-periments but the context is slightly differenthere The results described above clearly demon-strate that terrestrial bacteria and Archaea are ca-pable of surviving exposure to the conditions oftemperature and pressure prevalent at the sur-face of Europa and in the near-surface environ-ment with no special preparation They do not needto be particularly hardy organisms or even ex-tremophiles to do this Furthermore our resultsdemonstrate that at least some terrestrial mi-croorganisms can survive the cold and vacuumof deep interstellar space Thus the only remain-ing obstacle to panspermia is radiation Microbesensconced within crevices and fractures deep in-side silicate or icy celestial objects will be wellprotected against ultraviolet and charged particleirradiation only high-energy particles will posea significant threat

Figure 11 illustrates the near-infrared spectra


of D radiodurans at 200 120 and 20 K As withother heavily hydrated species such as epsomite(cf Dalton and Clark 1998 McCord et al 19982002 Dalton 2000) there is little change in thewater absorption features below 200 K This factmay simplify the search for life and other organicmaterial beyond the Solar System (Des Marais etal 2002 Pendleton and Allamandola 2002) be-cause only liquid and frozen organisms may needto be considered The search for evidence of lifebeyond the Solar System will not be limited toplanets at Earth-like temperatures and so thesemeasurements will prove useful Further workwill need to be done to ensure that other mi-croorganisms and materials do not exhibit spec-tral changes at low temperatures

The survival of even the relatively fragile Sshibatae sample has another implication Aspointed out earlier the Sulfolobus growthmedium is made up largely of sulfur com-pounds already considered to be likely compo-nents of the Europan surface and ocean (Table1) (McCord et al 1998 2002 Carlson et al 1999b2002 Kargel et al 2000 Dalton 2000) Even iflife is not found at Europa it is possible that ter-restrial organisms could be used to seed lifethere The ethical considerations of such activityare profound and naturally beyond the scope ofthis investigation However the results pre-sented here demonstrate that such conceptsmust be considered carefully in light of the po-

tential for accidental contamination of a possi-bly pristine world with terrestrial microbes

While it is highly unlikely that extant organ-isms might be found at the surface of Europa twoother possibilities remain One is that the radio-lytically processed detritus of endogenic mi-croorganisms emplaced at the surface may stillbe available for study As shown above such de-tritus may contain hydrated compounds aminoacids amides and other products of radiation-driven organic chemistry The reasonable as-sumption that Europan organisms might at leastpartially base their metabolisms upon abundantsulfur compounds suggests in turn that radiolyt-ically processed sulfur compounds derived fromliving material could exist at the surface

The other remaining possibility is that viableorganisms could persist trapped in the ice Basedon the resurfacing scenario presented above ifthe upper 13 m has been sufficiently processedto render it sterile then an astrobiological ex-plorer would need to excavate to at least 2 m orpossibly 5 m depth in order to find organismsthat while subject to extreme cold have not beenexposed to radiation in the geologically recentpast Furthermore given the ongoing removal ofsurface material by sputtering and micromete-orite impact processes relatively fresh materialmay be exposed over geologic time Lag depositsleft behind as more volatile molecules migrate tocolder surface areas could be expected to accu-

DALTON ET AL524

02

03

04

05

06

07

08 12 16 2 24

Ref

lect

ance

(p

lus

off

set)

Wavelength (mm)

200K

Deinococcus radiodurans at low temperature

120K

20KFIG 11 Spectra of D radiodurans at200 120 and 20 K Little variation is seenwith temperature what is seen here maybe experimental artifacts due to smallchanges in lighting and viewing geometry

mulate in the darker (and consequently warmer)disrupted regions Near-surface material that hasnot been subjected to the full surface exposureflux may retain greater complements of biologi-cally derived material

Noting the band depth of the 205-mm absorp-tion feature in the cryogenic S shibatae samplesto have been around 688 of continuum re-flectance (Table 2) and the absorption banddepth in the 72-NIMSEL observation to be264 a simple linear scaling may be used tocalculate a rough upper limit on surface abun-dances Based on the determination that our cen-trifuged samples had cell densities of around 5 3108 cellscm3 the cell density of entrained mi-croorganisms at Europa would be about 2 3 108

cellscm3 or 200 million cellsml This corre-sponds to about 02 mgcm3 of carbonaceous ma-terial Based on our survival rates we calculatethat within 5 m of the surface in dark disruptedterrain there could conceivably be as many as 4 31022 viable cellscm3 if the cells were as fragileas the S shibatae If on the other hand the em-placed cells were as hardy as the D radioduransit may be possible to find as many as 14 3 108

viable cellscm3 within 2ndash5 m of the surface indark deposits If microbes were emplaced at con-centrations of 106 cellscm3 which is typical forterrestrial saline lakes there could still be from 1cell per 46 l of ice to 50000 cellscm3 still viableEven if microbes were emplaced at a concentra-tion of 5 3 103 cellscm3 (Pierazzo and Chyba2002) this could still provide as many as 35 3103 viable cellscm3 if they are as tough as D ra-diodurans If however they are fragile this couldleave as few as 1 viable cell in 106 cm3 whichcould be difficult to search for It is important tobear in mind though that all it takes is one livingcell Though these calculated values are of ne-cessity highly speculative they provide a valu-able starting place for discussion of strategies forthe search for life on Europa

Future work

The results presented here suggest several ar-eas in which further work is needed The survivalof both biota and biotic materials subjected to par-ticle fluxes at energies temperatures and pres-sures relevant to Europa will be needed to an-swer a number of questions The survival of theamide bonds is of particular interest Should theamide bonds survive exposure to the Europan ra-

diation environment then they will provide apowerful biosignature that can be exploited in theevaluation of astrobiological potential at EuropaOn the other hand if the amide bonds are de-stroyed the spectrum of the resulting materialshould give a compelling spectral match to theobservations of reddish disrupted terrains Thishas profound implications

The nature of other compounds produced bysuch radiolysis is also of importance Analysis ofthis material could provide insights into the struc-ture of whatever the actual surface material is onEuropa based on spectral similarities The fam-ily of hydrates and hydrated salts in particularremains largely unexamined under controlledlaboratory conditions that simulate the Europanenvironment It may be possible for not only in-organic species but also biomolecules or com-pounds of a prebiotic nature to form hydrateseven when frozen perhaps under the influenceof ionizing radiation A great deal of work stillneeds to be done to assess the potential of vari-ous candidate materials to explain the observa-tions

A comprehensive search of the Galileo NIMSobservations may yet turn up more evidence ofnarrow absorption features that have hithertogone unnoticed Analysis of the Galileo mea-surements has been hampered by the high radi-ation noise present in the Europa flyby data andthe associated difficulties in interpreting the significance of small features compounded bythe lack of relevant laboratory measurementsProgress on both fronts could yield new insights

Finally future work should include plans fora return to Europa possibly with new spec-trometers having capabilities that can addressthe questions raised by the NIMS data An or-biting imaging spectrometer having 2- to 5-nmsampling and bandpass with a spatial resolutionup to 100 m should be capable of acquiring spec-tra of dark homogeneous surface units with suf-ficient signal to unequivocally identify narrowspectral features such as those common to bio-logical materials A landed mission capable ofpenetrating between 2 to 5 m (if not all the wayto the interior ocean) could collect samples andperform in situ analyses which would defini-tively answer many questions about the surfacecomposition While not a certainty there is alsothe possibility that it may be able to return evi-dence for biological activity in the subsurface ofthis enigmatic world


CONCLUSION

The comparison between observations of Eu-ropa from the Galileo NIMS instrument and lab-oratory measurements of microbe samples undertemperature and pressure conditions relevant tothe Europan surface provides unique constraintson possible evidence for extraterrestrial life In-frared absorption features due to water of hy-dration and amide bonds within cellular struc-tures may be useful as biosignatures in searchesfor past or present biota and biotic materials Ex-amination of the Galileo NIMS observations sug-gested tantalizing glimpses of this biosignaturehowever within the constraints of spectral reso-lution and signal-to-noise these NIMS observa-tions cannot be considered definitive evidence foreither extant or extinct life on Europa Howeverthe possibility of such evidence cannot be ruledout on the basis of available information eitherThe data do suggest that recoverable samples ofbiological material and even viable organisms ifpresent in these locations could persist within afew meters of the surface Additional laboratorywork at relevant Europan temperature and pres-sure conditions particularly on exposure of bio-logical material to particle bombardment andspectral analysis of other candidate surface ma-terials will be needed to address these questionsFurther analysis of the available NIMS data mayyield additional insights An orbiter with an ad-vanced imaging spectrometer should be able toresolve the narrow spectral features diagnostic ofbiological materials while a landed mission ca-pable of excavating several meters may be able tosample material that has not been substantiallyaltered by radiation and thus derive new knowl-edge of endogenic processes and their relation-ship to possible biological activity Future space-craft mission designs should take these resultsinto consideration in order to optimize their ca-pability to assess the astrobiological potential ofthis icy world

ACKNOWLEDGMENTS

Sincere thanks to Roger Clark and Ted Roushfor access to the cryogenic and spectrometer sys-tems and to Jack Lissauer and Jonathan Trent forhelpful discussions Also we are grateful to Wol-fram Zillig of the Max Planck Institute of Bio-chemistry for the S shibatae culture and Raymond

Kokaly of the US Geological Survey for the Cyani-dium algae spectrum which inspired much of thisproject We acknowledge helpful reviews byRobert Carlson and by an anonymous reviewerThis investigation was partly supported by the Na-tional Research Council Fellowship program Nomacroscopic organisms were physically harmedduring the preparation of this manuscript

ABBREVIATIONS

ASD Analytical Spectral Devices FWHM fullwidth at half-maximum NACREF NASA AmesCryogenic Reflectance Environment FacilityNIMS Near Infrared Imaging Spectrometer OSSEC Outer Solar System Environment Cham-ber SSI Solid State Imager

REFERENCES

Bolrsquoshakov AA Cruden BA Mogul R Rao MVVSSharma SP Khare BN and Meyyappan M (2003)Radio-frequency oxygen plasma as a sterilizationsource AIAA J (in press)

Borucki JG Khare B and Cruikshank DP (2002) Anew energy source for organic synthesis in Europarsquossurface ice J Geophys Res 107(E11) 5114 doi1010292002JE001841

Carlson RW Weissman PR Smythe WD MahoneyJC and the NIMS Science and Engineering Teams(1992) Near-Infrared Mapping Spectrometer experi-ment on Galileo Space Sci Rev 60 457ndash502

Carlson RW Anderson MS Johnson RE SmytheWD Hendrix AR Barth CA Soderblom LAHansen GB McCord TB Dalton JB Clark RNShirley JH Ocampo AC and Matson DL (1999a)Hydrogen peroxide on the surface of Europa Science283 2062ndash2064

Carlson RW Johnson RE and Anderson MS (1999b)Sulfuric acid on Europa and the radiolytic sulfur cycleScience 286 5437ndash5440

Carlson RW Anderson MS Johnson RE SchulmanMB and Yavrouian AH (2002) Sulfuric acid pro-duction on Europa the radiolysis of sulfur in water iceIcarus 157 456ndash463

Cassen P Reynolds RT and Peale SJ (1979) Is thereliquid water on Europa Geophys Res Lett 6 731ndash734

Chaban GM Huo WM and Lee TJ (2002) Theoreti-cal study of infrared and Raman spectra of hydratedmagnesium sulfate salts J Chem Phys 117 2532ndash2537

Chyba CF (2000) Energy for microbial life on EuropaNature 403 381ndash382

Chyba CF and Phillips C (2001) Possible ecosystemsand the search for Life on Europa Proc Natl Acad SciUSA 98 801ndash804

DALTON ET AL526

Clark RN (1993) SPECtrum Processing Routines UserrsquosManual Version 3 (program SPECPR) US GeologicalSurvey Open File Report 93ndash595 Available athttpspeclabcrusgsgov

Clark RN and McCord T B (1980) The Galilean satel-lites new near-infrared spectral reflectance measure-ments (065ndash25 mm) and a 0325ndash5 mm summary Icarus41 323

Clark RN and Roush TL (1984) Reflectance spec-troscopy quantitative analysis techniques for remotesensing applications J Geophys Res 89 6329ndash6340

Colthup NB Daly LH and Wiberley SE (1990) In-troduction to Infrared and Raman Spectroscopy 3rd editAcademic Press San Diego

Cooper JF Johnson RE Mauk BH Garrett HB andGehrels N (2001) Energetic ion and electron irradia-tion of the icy Galilean satellites Icarus 149 133ndash159

Cottrell TL (1958) The Strengths of Chemical Bonds But-terworths London

Crowley JK (1991) Visible and near-infrared (04ndash25mm) reflectance spectra of playa evaporite minerals JGeophys Res 96 16231ndash16240

Dalton JB (2000) Constraints on the surface compositionof Jupiterrsquos moon Europa based on laboratory andspacecraft data [PhD dissertation] University of Col-orado Boulder

Dalton JB (2002a) Detectability of potentially entrainedmicroorganisms at the surface of Europa [abstract]Proc Lunar Planet Sci Conf XXXIII 1555

Dalton JB (2002b) Cryogenic reflectance spectroscopy insupport of planetary missions In Proceedings of theNASA Laboratory Astrophysics Workshop edited by FSalama NASA CP-2002ndash211863 NASA Moffett FieldCA pp 241ndash244

Dalton JB (2003) Spectral behavior of hydrated sulfatesalts implications for Europa mission spectrometer de-sign Astrobiology (in press)

Dalton JB and Clark RN (1998) Laboratory spectra ofEuropa candidate materials at cryogenic temperatures[abstract] Bull Am Astron Soc 30 1081

Dalton JB and Clark RN (1999) Observational constraintson Europarsquos surface composition from Galileo NIMS data[abstract] Proc Lunar Planet Sci Conf XXX 2064

Darwent B deB (1970) Bond Dissociation Energies in Simple Molecules National Standard Reference DataSystem Publication 31 US National Bureau of Stan-dardsWashington DC

Delitsky ML and Lane AL (1997) Chemical schemesfor surface modification of icy satellites a road map JGeophys Res 102 16385

Delitsky ML and Lane AL (1998) Ice chemistry on theGalilean satellites J Geophys Res 103 31391ndash31403

Des Marais DJ Harwit MO Jucks KW Kasting JFLin DNC Lunine JI Schneider J Seager S TraubWA and Woolf NJ (2002) Remote sensing of plane-tary properties and biosignatures on extrasolar terres-trial planets Astrobiology 2 153ndash181

Eisenberg D and Kauzman W (1969) The Structure andProperties of Water Oxford University Press London

Fanale FP Li YH Decarlo E Domergue-Schmidt N

Sharma SK Horton K Granahan JC and theGalileo NIMS Team (1998) Laboratory simulation of thechemical evolution of Europarsquos aqueous phase [ab-stract] Proc Lunar Planet Sci XXIX 1248

Fanale FP Granahan JC McCord TB Hansen G Hibbitts CA Carlson R Matson D Ocampo AKamp L Smythe W Leader F Mehlman R GreeleyR Sullivan R Geissler P Barth C Hendrix A ClarkB Helfenstein P Veverka J Belton M J S Becker KBecker T and the Galileo NIMS SSI UVS InstrumentTeams (1999) Galileorsquos multi-instrument spectral view ofEuroparsquos surface composition Icarus 139 179ndash188

Fanale FP Granahan JC Greeley R Pappalardo RHead J Shirley J Carlson R Hendrix A Moore JMcCord TB and Belton M (2000) Tyre and PwyllGalileo orbital remote sensing of mineralogy versusmorphology at two selected sites on Europa J GeophysRes 105 22647ndash22657

Figueredo PH Chuang FC Rathbun J Kirk RL andGreeley R (2002) Geology and origin of Europarsquos ldquoMit-tenrdquo feature (Murias Chaos) J Geophys Res 107 doi1010292001JE001591

Fink U and Sill G (1982) The infrared spectral proper-ties of frozen volatiles In Comets edited by L Wilken-ing University of Arizona Press Tucson pp 164ndash202

Fredrickson JK Kostandarithes HW Li SW PlymaleAE and Daly MJ (2000) Reduction of Fe(III) Cr(VI)U(VI) and Tc(VII) by Deinococcus radiodurans R1 ApplEnviron Microbiol 66 2006ndash2011

Freshney RI (2000) Culture of Animal Cells A Manual ofBasic Technique Wiley New York

Gaffey SJ McFadden LA Nash D and Pieters CM(1993) Ultraviolet visible and near-infrared reflectancespectroscopy laboratory spectra of geologic materialsIn Remote Geochemical Analysis edited by CM Pietersand PAJ Englert Cambridge University Press Cam-bridge pp 43ndash77

Garrison WM (1968) Radiation chemistry of organo-nitrogen compounds Curr Top Radiat Res 4 45ndash90

Garrison WM (1987) Reaction mechanisms in the radi-olysis of peptides polypeptides and proteins ChemRev 87 381ndash398

Goetz AF Heidebrecht KB Kindel B and BoardmanJW (1998) Using ground spectral irradiance for modelcorrection of AVIRIS data In Summaries of the SeventhAnnual JPL Airborne Earth Science Workshop JPL Publi-cation No 97ndash21 Jet Propulsion Laboratory PasadenaCA pp 159ndash168

Granahan JC Fanale FP Carlson R Kamp L Mat-son D Ocampo A Smythe W Greeley R GeisslerP Moore J M Belton M and the Galileo SSI andNIMS Teams (1998) Galileo remote sensing composi-tional studies of the Tyre Macula region of Europa [ab-stract] Proc Lunar Planet Sci XXIX 1291

Greenberg R Hoppa GV Tufts BR Geissler P Ri-ley J and Kandel S (1999) Chaos on Europa Icarus141 263ndash286

Grogan DW (1989) Phenotypic characterization of thearchaebacterial genus Sulfolobus comparison of fivewild-type strains J Bacteriol 171 6710ndash6719


Grundy WM and Schmitt B (1998) The temperature-dependent near-infrared absorption spectrum of hexag-onal H2O ice J Geophys Res 103 25809ndash25822

Hay RJ (1978) Preservation of cell culture stocks in liq-uid nitrogen TCA Manual 4 787ndash790

Head JW and Pappalardo RT (1999) Brine mobiliza-tion during lithospheric heating on Europa implica-tions for formation of chaos terrain lenticula textureand color variations J Geophys Res 104 27143ndash27156

Herzberg G (1991) Molecular Spectra and Molecular Struc-ture II Krieger Malabar FL

Hobbs PV (1974) Ice Physics Oxford University PressLondon

Hochanadel CJ (1960) Radiation chemistry of water InComparative Effects of Radiation edited by M Burton JSKirby-Smith and JL Magee Wiley New York pp151ndash189

Hudson RL and Moore MH (2001) Radiation chemi-cal alterations in solar system ices an overview J Geo-phys Res 106 33275ndash33284

Hunt GR and Ashley RP (1979) Spectra of altered rocksin the visible and near infrared Econ Geol 741613ndash1629

Hunt GR Salisbury JW and Lenhoff C (1971) Visibleand near-infrared spectra of minerals and rocks IV Sul-phides and sulphates Mod Geol 3 1ndash14

Hunt GR Salisbury JW and Lenhoff CJ (1973) Visi-ble and near-infrared spectroscopy of minerals androcks VI Additional silicates Mod Geol 4 85ndash106

Ip W-H Williams DJ McEntire RW and Mauk BH(1998) Ion sputtering and surface erosion at EuropaGeophys Res Lett 25 829ndash832

Jamieson CS and Dalton JB (2002) A summary of lab-oratory spectra for ices of planetary interest Bull AmAstron Soc 34 3707

Joesten MD and Schaad LJ (1974) Hydrogen BondingMarcel Dekker New York

Johnson RE and Quickenden TI (1997) Photolysis andradiolysis of water ice on outer solar system bodies JGeophys Res 102 10985ndash10996

Kargel JS (1998) The salt of Europa Science 2801211ndash1212

Kargel JS (1991) Brine volcanism and the interior struc-ture of asteroids and icy satellites Icarus 94 368ndash390

Kargel JS Kaye JZ Head JW III Marion GMSassen R Crowley JK Prieto-Ballesteros O GrantSA and Hogenboom DL (2000) Europarsquos crust andocean origin composition and the prospects for lifeIcarus 148 226ndash265

Kempe S and Kazmierczak J (2002) Biogenesis and earlylife on Earth and Europa favored by an alkaline oceanAstrobiology 2 123ndash130

Kivelson MG Khurana KK Russell CT Volwerk MWalker RJ and Zimmer C (2000) Galileo magne-tometer measurements a stronger case for a subsurfaceocean at Europa Science 289 1380ndash1343

Kminek G and Bada JL (2001) The effect of sterilizinggamma radiation on amino acids [abstract] Proc LunarPlanet Sci XXXII 1918

Kvenvolden KA Lawless JG Pering K Peterson EFlores J Ponnamperuma C Kaplan IR and Moore

C (1970) Evidence for extraterrestrial amino acids andhydrocarbons in the Murchison meteorite Nature 228923ndash926

Lane AL Nelson RM and Matson DL (1981) Evi-dence for sulphur implantation in Europarsquos UV ab-sorption Nature 292 38ndash39

Lange CC Wackett LP Minton KW and Daly MJ(1998) Engineering a recombinant Deinococcus radiodu-rans for organopollutant degradation in radioactivemixed waste environments Nat Biotechnol 16 929ndash933

March J (1968) Advanced Organic Chemistry ReactionsMechanisms and Structure McGraw-Hill New York

Marion GM and Farren RE (1999) Mineral solubili-ties in the Na-K-Mg-Ca-Cl-SO4-H2O system a re-eval-uation of the sulfate chemistry in the Spencer-Moslashller-Weare model Geochim Cosmochim Acta 63 1305ndash1318

Mattimore V and Battista JR (1996) Radioresistance ofDeinococcus radiodurans functions necessary to surviveionizing radiation are also necessary to survive pro-longed desiccation J Bacteriol 178 633ndash637

Matsumoto A Hayashi N and Lichtin NN (1970) 60Cogamma-radiolysis of liquid formamide gaseous andhigh-boiling stable products Radiat Res 41 299ndash311

Mazur P (1980) Limits to life at low temperatures and atreduced water contents and water activities Origins Life10 137ndash159

McCord TB Hansen GB Fanale FP Carlson RWMatson DL Johnson TV Smythe WD CrowleyJK Martin PD Ocampo A Hibbitts CA Grana-han JC and the NIMS Team (1998) Salts on Europadetected by Galileorsquos Near Infrared Mapping Spec-trometer Science 280 1242ndash1245

McCord TB Hansen GB Matson DL Johnson TVCrowley JK Fanale FP Carlson RW SmytheWD Martin PD Hibbitts CA Granahan JCOcampo A and the NIMS team (1999a) Hydrated saltminerals on Europarsquos surface from the Galileo near-infrared mapping spectrometer (NIMS) investigation JGeophys Res 104 11824ndash11852

McCord TB Hansen GB Shirley JH and CarlsonRW (1999b) Discussion of the 104-mm water ice ab-sorption in the Europa NIMS spectra and a new NIMScalibration J Geophys Res 104 27157ndash27162

McCord TB Orlando TM Teeter G Hansen GBSieger MT Petrik NG and van Keulen L (2001)Thermal and radiation stability of the hydrated saltminerals epsomite mirabilite and natron under Eu-ropa environmental conditions J Geophys Res 106(E2)3311ndash3320

McCord TB Teeter G Hansen GB Sieger MT andOrlando TM (2002) Brines exposed to Europa sur-face conditions J Geophys Res 107 doi 1010292000JE001453

McKinnon WB (1999) Convective instability in Europarsquosfloating ice shell Geophys Res Lett 26 951ndash954

McKinnon WB and Shock EL (2002) Ocean karmawhat goes around comes around (or does it) [abstract]Proc Lunar Planet Sci Conf XXXIII 2181

Mogul R Bolrsquoshakov AA Chan SL Stevens RMKhare BN Meyyappan M and Trent JD (2003) Im-

DALTON ET AL528

pact of low temperature plasmas on Deinococcus radio-durans and biomolecules Biotechnol Prog 19 776ndash783doi 101021bp025665e

Moore MH (1984) Studies of proton-irradiated SO2

at low temperatures implications for Io Icarus 59114ndash128

Nash DB and Fanale FP (1977) Iorsquos surface composi-tion based on reflectance spectra of sulfursalt mixturesand proton-irradiation experiments Icarus 31 40ndash80

Noll KS Weaver HA and Gonnella AM (1995) The albedo spectrum of Europa from 2200 to 3300Angstroms J Geophys Res 100 19057ndash19059

Novak A (1974) Hydrogen bonding in solids Correla-tion of spectroscopic and crystallographic data StructBonding 18 177ndash212

Ockman N (1957) The infrared-spectra and Raman-spectra of single crystals of ordinary ice [PhD disser-tation] University of Michigan Ann Arbor

Orlando TM (2002) Electron- and vacuum ultravioletphoton-induced weathering of outer solar system sur-faces [abstract P52C-07] In EOS Transactions of theAmerican Geophysical Union 2002 Fall Meeting Suppl 83P52C-07

Pappalardo RT Belton MJS Breneman HH CarrMH Chapman CR Collins GC Denk T FagentsS Geissler PE Giese B Greeley R Greenberg RHead JW Helfenstein P Hoppa G Kadel SDKlaasen KP Klemaszewski JE Magee K McEwenAS Moore JM Moore WB Neukum G PhillipsCB Prockter LM Schubert G Senske DA Sulli-van RJ Tufts BR Turtle EP Wagner R andWilliams KK (1999) Does Europa have a subsurfaceocean Evaluation of the geological evidence J Geo-phys Res 104 24015ndash24056

Paranicas C Carlson RW and Johnson RE (2001)Electron bombardment of Europa Geophys Res Lett 28673ndash676

Paranicas C Mauk BH Ratliff JM Cohen C andJohnson RE (2002) The ion environment near Europaand its role in surface energetics Geophys Res Lett 2918-1ndash18-4 doi 1010292001GL014127

Pauling L (1935) The structure and entropy of ice andother crystals with some randomness of atomicarrangement J Am Chem Soc 57 2680ndash2684

Pauling L (1960) The Nature of the Chemical Bond 3rd editCornell University Press Ithaca NY

Pendleton YJ and Allamandola LJ (2002) The organicrefractory material in the diffuse interstellar mediummid-infrared spectroscopic constraints Astrophys J138 75ndash98

Pierazzo E and Chyba CF (1999) Amino acid survivalin large cometary impacts Meteorit Planet Sci 34909ndash918

Pierazzo E and Chyba CF (2002) Cometary delivery ofbiogenic elements to Europa Icarus 157 120ndash127

Pilcher CB Ridgway ST and McCord TB (1972)Galilean satellites identification of water frost Science178 1087ndash1089

Prockter LM Head JW Pappalardo RT SullivanRJ Clifton AE Giese B Wagner R and NeukumG (2002) Morphology of Europan bands at high reso-

lution a mid-ocean ridge-type rift mechanism J Geo-phys Res 107 doi 1010292000JE001458

Prockter LM and Pappalardo RT (2000) Folds on Eu-ropa implications for crustal cycling and accommoda-tion of extension Science 289 941ndash943

Reynolds RT Squyres SW Colburn DS and McKayCP (1983) On the habitability of Europa Icarus 56246ndash254

Rivkina EM Friedmann EI McKay CP and Gilichin-sky DA (2000) Metabolic activity of permafrost bac-teria below the freezing point Appl Environ Microbiol66 3230ndash3233

Rothschild LJ and Mancinelli RL (2001) Life in extremeenvironments Nature 409 1092ndash1101

Roush TL (2001) Physical state of ices in the outer solarsystem J Geophys Res 106 33315ndash33323

Sagan C Thompson WR Carlson R Gurnett D andHord C (1993) A search for life from the Galileo space-craft Nature 365 715

Schulze-Makuch D Irwin LN and Guan H (2002)Search parameters for the remote detection of extrater-restrial life Planet Space Sci 50 675ndash683

Spaun NA and Head JW (2001) A model of Europarsquoscrustal structure recent Galileo results and implicationsfor an ocean J Geophys Res 106 7567ndash7576

Spaun NA Head JW Collins GC Prockter LM andPappalardo RT (1998) Conamara Chaos region Eu-ropa reconstruction of mobile ice polygons GeophysRes Lett 25 4277ndash4280

Squyres SW Reynolds RT and Cassen PM (1983)Liquid water and active resurfacing on Europa Nature301 225ndash226

Stevenson D (2000) Europarsquos oceanmdashthe case strength-ens Science 289 1305ndash1307

Whitman WB Coleman DC and Wiebe W J (1998)Prokaryotes the unseen majority Proc Natl Acad SciUSA 95 6578ndash6583

Venkateswaran A McFarlan SC Ghosal D MintonKW Vasilenko A Makarova K Wackett LP andDaly MJ (2000) Physiologic determinants of radiationresistance in Deinococcus radiodurans Appl Environ Mi-crobiol 66 2620ndash2626

Zahnle K Levison H and Dones L (1998) Crateringrates on the Galilean satellites Icarus 136 202ndash222

Zolotov MY and Shock EL (2001) The composition andstability of salts on the surface of Europa and theiroceanic origin J Geophys Res 106 32815ndash31827

Zolotov MY and Shock EL (2003) Energy for biologicsulfate reduction in a hydrothermally formed ocean on Europa J Geophys Res 108(E4) 5022 doi 1010292002JE001966

Address reprint requests toDr J Brad Dalton

SETI Institute MS 245-3NASA Ames Research CenterMoffett Field CA 94035-1000

E-mail daltonmailarcnasagov


et al 1983 Sagan et al 1993 Fanale et al 1998Chyba and Phillips 2001 Rothschild and Man-cinelli 2001 Zolotov and Shock 2003)

Because the putative ocean remains inaccessi-ble beneath Europarsquos icy crust infrared remotesensing of the surface materials has been used toinfer details about the interior composition If infact the surface composition is indicative of inte-rior composition studies of surface materialsmay provide insight into biological and prebioticprocesses (Squyres et al 1983 McCord et al1999a Kargel et al 2000) The diagnostic spectralcharacteristics of specific biogenic materials mayprovide unique biosignatures (Schulze-Makuchet al 2002) which can be tested for in the searchfor extraterrestrial life Infrared detection of evi-dence for life requires that (1) material of biolog-ical origin be emplaced at the surface (2) identi-fiable infrared absorption features diagnostic ofthis material survive exposure to the hostile spaceenvironment and (3) spectra of sufficient spectralresolution and signal-to-noise ratio be acquiredat spatial scales capable of isolating regions con-taining the material of interest This paper appliesthese criteria to the examination of recent space-craft and laboratory data concerning the spectraof Europa and those of terrestrial microorgan-isms

The surface of Europa is composed mostly ofwater ice (Pilcher et al 1972 Clark and McCord1980) More recent ground-based and spacecraftobservations have indicated the presence of othermaterials as well Cometary and meteoric infallare expected to provide several elemental andmolecular species over the age of the Solar Sys-tem including biogenic elements (Zahnle et al1998 Pierazzo and Chyba 2002) Energetic parti-cle bombardment associated with Jupiterrsquos mag-netosphere not only delivers elements such as HS and O but also drives complex chemistry thatis expected to produce a number of additionalcompounds (Delitsky and Lane 1997 1998Cooper et al 2001 Carlson et al 2002) Observedspecies include O2 SO2 CO2 and H2O2 (Lane etal 1981 Noll et al 1995 Carlson et al 1999ab)Current models suggest the presence of sulfuricacid octahydrate and other sulfur compounds(Carlson et al 1999b) simple organics such asformaldehyde (Delitsky and Lane 1998 Chyba2000) and hydrated sulfate and carbonate salts(Kargel 1991 1998 McCord et al 1998 1999ab2002 Kargel et al 2000) While no single mater-ial completely explains the Galileo observations

diagnostic infrared absorption features haveproven useful for constraining the abundances ofseveral compounds (Dalton 2000)

Water ice and frost have well-characterizedsymmetric near-infrared absorption features at15 and 20 mm (Ockman 1957 Herzberg 1991Gaffey et al 1993) These are evident in theGalileo Near-Infrared Imaging Spectrometer(NIMS) Europa spectrum of icy terrain in Fig 1(bottom curve) Other regions of the surface dis-play distorted asymmetric features at these po-sitions as indicated by the top curve The pres-ence of these distorted absorption features intelescopic and spacecraft observations of Europa(Clark and McCord 1980 Dalton 2000) forms thebasis of the argument for hydrated salts on thesurface (McCord et al 1998 1999a) These fea-tures are clearly due to water of hydration (Paul-ing 1935) though the host molecule is less cer-tain (Gaffey et al 1993 Carlson et al 1999b

DALTON ET AL506

0

01

02

03

04

05

06

07

08

08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Icy Plains

Dark terrain

FIG 1 Galileo NIMS spectra of both icy and dark sur-face units on Europa Lower curve is a 26-NIMSEL av-erage from an icy region illustrating the broad 20-mmabsorption feature indicating pure H2O ice Upper curveis an average of 18 NIMSELS corresponding to dark red-dish-brown terrain This curve exhibits the distorted andasymmetric 20-mm absorption bands typical of boundwater such as water of hydration Both averages were ex-tracted from contiguous portions of the G1ENNHILAT01observation (Dalton 2000) Overlapping data segmentsat every 12th wavelength position are instrumental arti-facts created by stepping the grating to acquire readingsat each detector the resulting segments are compiled tocreate each spectrum (Carlson et al 1992) with redun-dancy supplied by the overlapping wavelengths















DALTON ET AL508


























DALTON ET AL510








RESULTS











DALTON ET AL512

0

02

04

06

08

1

08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Escherichia coli


Sulfolobus shibatae


N

H

CC

CO







0

02

04

06

08

1

12

08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Lipid

DNA

Protein

Carbohydrate

C-H



08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Europa

Hexahydrite

Bloedite

Sulfolobus Shibatae

H2SO4 8H2O






DALTON ET AL514

18 19 2 21 22 23

05

1

15

Co

nti

nu

um

-co

rrec

ted

ref

lect

ance

Wavelength (mm)

Europa

Hexahydrite

Bloedite

H2SO4 8H2O

Sulfolobus Shibatae

a b c





















DALTON ET AL516


210 W

20 N

10 N

NIMS E11ENCYCLOD


0

005

01

015

02

18 19 2 21 22 23


Ref

lect

ance

Wavelength (mm)

72 NIMSEL AVERAGE

21 NIMSEL










DISCUSSION


DALTON ET AL518





05

06

07

08

09

1

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)









DALTON ET AL520

0

02

04

06

08

1

12

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)


NIMS E11 ENCYCLOD01













DALTON ET AL522




C C 498 6 21C H 335 6 4C N 2782N H 372 6 8














DALTON ET AL524

02

03

04

05

06

07

08 12 16 2 24

Ref

lect

ance

(p

lus

off

set)

Wavelength (mm)

200K


120K






Future work







CONCLUSION


ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES











DALTON ET AL526








































































DALTON ET AL528






















































DALTON ET AL508


























DALTON ET AL510








RESULTS











DALTON ET AL512

0

02

04

06

08

1

08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Escherichia coli


Sulfolobus shibatae


N

H

CC

CO







0

02

04

06

08

1

12

08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Lipid

DNA

Protein

Carbohydrate

C-H



08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Europa

Hexahydrite

Bloedite

Sulfolobus Shibatae

H2SO4 8H2O






DALTON ET AL514

18 19 2 21 22 23

05

1

15

Co

nti

nu

um

-co

rrec

ted

ref

lect

ance

Wavelength (mm)

Europa

Hexahydrite

Bloedite

H2SO4 8H2O

Sulfolobus Shibatae

a b c





















DALTON ET AL516


210 W

20 N

10 N

NIMS E11ENCYCLOD


0

005

01

015

02

18 19 2 21 22 23


Ref

lect

ance

Wavelength (mm)

72 NIMSEL AVERAGE

21 NIMSEL










DISCUSSION


DALTON ET AL518





05

06

07

08

09

1

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)









DALTON ET AL520

0

02

04

06

08

1

12

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)


NIMS E11 ENCYCLOD01













DALTON ET AL522




C C 498 6 21C H 335 6 4C N 2782N H 372 6 8














DALTON ET AL524

02

03

04

05

06

07

08 12 16 2 24

Ref

lect

ance

(p

lus

off

set)

Wavelength (mm)

200K


120K






Future work







CONCLUSION


ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES











DALTON ET AL526








































































DALTON ET AL528

































































DALTON ET AL510








RESULTS











DALTON ET AL512

0

02

04

06

08

1

08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Escherichia coli


Sulfolobus shibatae


N

H

CC

CO







0

02

04

06

08

1

12

08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Lipid

DNA

Protein

Carbohydrate

C-H



08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Europa

Hexahydrite

Bloedite

Sulfolobus Shibatae

H2SO4 8H2O






DALTON ET AL514

18 19 2 21 22 23

05

1

15

Co

nti

nu

um

-co

rrec

ted

ref

lect

ance

Wavelength (mm)

Europa

Hexahydrite

Bloedite

H2SO4 8H2O

Sulfolobus Shibatae

a b c





















DALTON ET AL516


210 W

20 N

10 N

NIMS E11ENCYCLOD


0

005

01

015

02

18 19 2 21 22 23


Ref

lect

ance

Wavelength (mm)

72 NIMSEL AVERAGE

21 NIMSEL










DISCUSSION


DALTON ET AL518





05

06

07

08

09

1

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)









DALTON ET AL520

0

02

04

06

08

1

12

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)


NIMS E11 ENCYCLOD01













DALTON ET AL522




C C 498 6 21C H 335 6 4C N 2782N H 372 6 8














DALTON ET AL524

02

03

04

05

06

07

08 12 16 2 24

Ref

lect

ance

(p

lus

off

set)

Wavelength (mm)

200K


120K






Future work







CONCLUSION


ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES











DALTON ET AL526








































































DALTON ET AL528















































RESULTS











DALTON ET AL512

0

02

04

06

08

1

08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Escherichia coli


Sulfolobus shibatae


N

H

CC

CO







0

02

04

06

08

1

12

08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Lipid

DNA

Protein

Carbohydrate

C-H



08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Europa

Hexahydrite

Bloedite

Sulfolobus Shibatae

H2SO4 8H2O






DALTON ET AL514

18 19 2 21 22 23

05

1

15

Co

nti

nu

um

-co

rrec

ted

ref

lect

ance

Wavelength (mm)

Europa

Hexahydrite

Bloedite

H2SO4 8H2O

Sulfolobus Shibatae

a b c





















DALTON ET AL516


210 W

20 N

10 N

NIMS E11ENCYCLOD


0

005

01

015

02

18 19 2 21 22 23


Ref

lect

ance

Wavelength (mm)

72 NIMSEL AVERAGE

21 NIMSEL










DISCUSSION


DALTON ET AL518





05

06

07

08

09

1

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)









DALTON ET AL520

0

02

04

06

08

1

12

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)


NIMS E11 ENCYCLOD01













DALTON ET AL522




C C 498 6 21C H 335 6 4C N 2782N H 372 6 8














DALTON ET AL524

02

03

04

05

06

07

08 12 16 2 24

Ref

lect

ance

(p

lus

off

set)

Wavelength (mm)

200K


120K






Future work







CONCLUSION


ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES











DALTON ET AL526








































































DALTON ET AL528














































DALTON ET AL512

0

02

04

06

08

1

08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Escherichia coli


Sulfolobus shibatae


N

H

CC

CO







0

02

04

06

08

1

12

08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Lipid

DNA

Protein

Carbohydrate

C-H



08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Europa

Hexahydrite

Bloedite

Sulfolobus Shibatae

H2SO4 8H2O






DALTON ET AL514

18 19 2 21 22 23

05

1

15

Co

nti

nu

um

-co

rrec

ted

ref

lect

ance

Wavelength (mm)

Europa

Hexahydrite

Bloedite

H2SO4 8H2O

Sulfolobus Shibatae

a b c





















DALTON ET AL516


210 W

20 N

10 N

NIMS E11ENCYCLOD


0

005

01

015

02

18 19 2 21 22 23


Ref

lect

ance

Wavelength (mm)

72 NIMSEL AVERAGE

21 NIMSEL










DISCUSSION


DALTON ET AL518





05

06

07

08

09

1

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)









DALTON ET AL520

0

02

04

06

08

1

12

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)


NIMS E11 ENCYCLOD01













DALTON ET AL522




C C 498 6 21C H 335 6 4C N 2782N H 372 6 8














DALTON ET AL524

02

03

04

05

06

07

08 12 16 2 24

Ref

lect

ance

(p

lus

off

set)

Wavelength (mm)

200K


120K






Future work







CONCLUSION


ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES











DALTON ET AL526








































































DALTON ET AL528













































0

02

04

06

08

1

12

08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Lipid

DNA

Protein

Carbohydrate

C-H



08 12 16 2 24

Sca

led

Ref

lect

ance

Wavelength (mm)

Europa

Hexahydrite

Bloedite

Sulfolobus Shibatae

H2SO4 8H2O






DALTON ET AL514

18 19 2 21 22 23

05

1

15

Co

nti

nu

um

-co

rrec

ted

ref

lect

ance

Wavelength (mm)

Europa

Hexahydrite

Bloedite

H2SO4 8H2O

Sulfolobus Shibatae

a b c





















DALTON ET AL516


210 W

20 N

10 N

NIMS E11ENCYCLOD


0

005

01

015

02

18 19 2 21 22 23


Ref

lect

ance

Wavelength (mm)

72 NIMSEL AVERAGE

21 NIMSEL










DISCUSSION


DALTON ET AL518





05

06

07

08

09

1

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)









DALTON ET AL520

0

02

04

06

08

1

12

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)


NIMS E11 ENCYCLOD01













DALTON ET AL522




C C 498 6 21C H 335 6 4C N 2782N H 372 6 8














DALTON ET AL524

02

03

04

05

06

07

08 12 16 2 24

Ref

lect

ance

(p

lus

off

set)

Wavelength (mm)

200K


120K






Future work







CONCLUSION


ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES











DALTON ET AL526








































































DALTON ET AL528













































DALTON ET AL514

18 19 2 21 22 23

05

1

15

Co

nti

nu

um

-co

rrec

ted

ref

lect

ance

Wavelength (mm)

Europa

Hexahydrite

Bloedite

H2SO4 8H2O

Sulfolobus Shibatae

a b c





















DALTON ET AL516


210 W

20 N

10 N

NIMS E11ENCYCLOD


0

005

01

015

02

18 19 2 21 22 23


Ref

lect

ance

Wavelength (mm)

72 NIMSEL AVERAGE

21 NIMSEL










DISCUSSION


DALTON ET AL518





05

06

07

08

09

1

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)









DALTON ET AL520

0

02

04

06

08

1

12

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)


NIMS E11 ENCYCLOD01













DALTON ET AL522




C C 498 6 21C H 335 6 4C N 2782N H 372 6 8














DALTON ET AL524

02

03

04

05

06

07

08 12 16 2 24

Ref

lect

ance

(p

lus

off

set)

Wavelength (mm)

200K


120K






Future work







CONCLUSION


ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES











DALTON ET AL526








































































DALTON ET AL528



























































DALTON ET AL516


210 W

20 N

10 N

NIMS E11ENCYCLOD


0

005

01

015

02

18 19 2 21 22 23


Ref

lect

ance

Wavelength (mm)

72 NIMSEL AVERAGE

21 NIMSEL










DISCUSSION


DALTON ET AL518





05

06

07

08

09

1

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)









DALTON ET AL520

0

02

04

06

08

1

12

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)


NIMS E11 ENCYCLOD01













DALTON ET AL522




C C 498 6 21C H 335 6 4C N 2782N H 372 6 8














DALTON ET AL524

02

03

04

05

06

07

08 12 16 2 24

Ref

lect

ance

(p

lus

off

set)

Wavelength (mm)

200K


120K






Future work







CONCLUSION


ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES











DALTON ET AL526








































































DALTON ET AL528









































210 W

20 N

10 N

NIMS E11ENCYCLOD


0

005

01

015

02

18 19 2 21 22 23


Ref

lect

ance

Wavelength (mm)

72 NIMSEL AVERAGE

21 NIMSEL










DISCUSSION


DALTON ET AL518





05

06

07

08

09

1

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)









DALTON ET AL520

0

02

04

06

08

1

12

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)


NIMS E11 ENCYCLOD01













DALTON ET AL522




C C 498 6 21C H 335 6 4C N 2782N H 372 6 8














DALTON ET AL524

02

03

04

05

06

07

08 12 16 2 24

Ref

lect

ance

(p

lus

off

set)

Wavelength (mm)

200K


120K






Future work







CONCLUSION


ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES











DALTON ET AL526








































































DALTON ET AL528















































DISCUSSION


DALTON ET AL518





05

06

07

08

09

1

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)









DALTON ET AL520

0

02

04

06

08

1

12

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)


NIMS E11 ENCYCLOD01













DALTON ET AL522




C C 498 6 21C H 335 6 4C N 2782N H 372 6 8














DALTON ET AL524

02

03

04

05

06

07

08 12 16 2 24

Ref

lect

ance

(p

lus

off

set)

Wavelength (mm)

200K


120K






Future work







CONCLUSION


ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES











DALTON ET AL526








































































DALTON ET AL528












































05

06

07

08

09

1

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)









DALTON ET AL520

0

02

04

06

08

1

12

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)


NIMS E11 ENCYCLOD01













DALTON ET AL522




C C 498 6 21C H 335 6 4C N 2782N H 372 6 8














DALTON ET AL524

02

03

04

05

06

07

08 12 16 2 24

Ref

lect

ance

(p

lus

off

set)

Wavelength (mm)

200K


120K






Future work







CONCLUSION


ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES











DALTON ET AL526








































































DALTON ET AL528














































DALTON ET AL520

0

02

04

06

08

1

12

18 19 2 21 22 23

Co

nti

nu

um

-co

rrec

ted

Ref

lect

ance

Wavelength (mm)


NIMS E11 ENCYCLOD01













DALTON ET AL522




C C 498 6 21C H 335 6 4C N 2782N H 372 6 8














DALTON ET AL524

02

03

04

05

06

07

08 12 16 2 24

Ref

lect

ance

(p

lus

off

set)

Wavelength (mm)

200K


120K






Future work







CONCLUSION


ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES











DALTON ET AL526








































































DALTON ET AL528



















































DALTON ET AL522




C C 498 6 21C H 335 6 4C N 2782N H 372 6 8














DALTON ET AL524

02

03

04

05

06

07

08 12 16 2 24

Ref

lect

ance

(p

lus

off

set)

Wavelength (mm)

200K


120K






Future work







CONCLUSION


ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES











DALTON ET AL526








































































DALTON ET AL528




















































DALTON ET AL524

02

03

04

05

06

07

08 12 16 2 24

Ref

lect

ance

(p

lus

off

set)

Wavelength (mm)

200K


120K






Future work







CONCLUSION


ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES











DALTON ET AL526








































































DALTON ET AL528












































Future work







CONCLUSION


ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES











DALTON ET AL526








































































DALTON ET AL528








































CONCLUSION


ACKNOWLEDGMENTS



ABBREVIATIONS


REFERENCES











DALTON ET AL526








































































DALTON ET AL528















































































































DALTON ET AL528








































methodologies and techniques for detecting extraterrestrial...

Documents