ASTROBIOLOGYVolume 8, Number 3, 2008© Mary Ann Liebert, Inc.DOI: 10.1089/ast.2008.0759

News and Views

Science Priorities for Mars Sample Return

The MEPAG Next Decade Science Analysis Group

Table of Contents

I. Executive Summary 491II. Introduction 493

III. Evaluation Process 494IV. Scientific Objectives of MSR 494

IV-A. History, current context of MSR’s scientific objectives 494IV-B. Possible scientific objectives for a next decade MSR 496

V. Samples Required to Achieve the Scientific Objectives 499V-A. Sedimentary materials rock suite 500V-B. Hydrothermal rock suite 500V-C. Low-temperature altered rock suite 501V-D. Igneous rock suite 501V-E. Regolith 502V-F. Polar ice 503V-G. Atmospheric gas 504V-H. Dust 505V-I. Depth-resolved suite 505V-J. Other 507

VI. Factors that Would Affect the Scientific Value of the Returned Samples 507VI-A. Sample size 507VI-B. Number of samples 512VI-C. Sample encapsulation 514VI-D. Diversity of the returned collection 515VI-E. In situ measurements for sample selection and documentation of field context 515VI-F. Surface operations 516VI-G. Sample acquisition system priorities 517VI-H. Temperature 517VI-I. Planning considerations involving the MSL/ExoMars caches 518VI-J. Planetary protection 521

489

The MEPAG Next Decade Science Analysis Group (ND-SAG): Lars E. Borg (co-chair), David J. Des Marais (co-chair), David W. Beaty,Oded Aharonson, Steven A. Benner, Donald D. Bogard, John C. Bridges, Charles J. Budney, Wendy M. Calvin, Benton C. Clark, JenniferL. Eigenbrode, Monica M. Grady, James W. Head, Sidney R. Hemming, Noel W. Hinners, Victoria Hipkin, Glenn J. MacPherson, LuciaMarinangeli, Scott M. McLennan, Harry Y. McSween, Jeffrey E. Moersch, Kenneth H. Nealson, Lisa M. Pratt, Kevin Righter, Steve W. Ruff,Charles K. Shearer, Andrew Steele, Dawn Y. Sumner, Steven J. Symes, Jorge L. Vago, and Frances Westall.

With input from the following experts:MEPAG Goal I. Marion Anderson (Monash University, Australia), Mike Carr (USGS, retired), Pamela Conrad (JPL), Danny Glavin

(GSFC), Tori Hoehler (NASA/ARC), Linda Jahnke (NASA/ARC), Paul Mahaffy (GSFC), Bruce Schaefer (Monash University, Australia),Andy Tomkins (Monash University, Australia), Aaron Zent (ARC);

MEPAG Goal II. Steve Bougher (University of Michigan), Shane Byrne (University of Arizona), Dorthe Dahl-Jensen (University of Copen-hagen), John Eiler (Caltech), Walt Engelund (LaRC), James Farquahar (University of Maryland), David Fernandez-Remolar (CAB, Spain),Kate Fishbaugh (Smithsonian), David Fisher (Geological Survey of Canada), Veronika Heber (Switzerland), Mike Hecht (JPL), Joel Hurowitz(JPL), Christine Hvidberg (University of Copenhagen), Bruce Jakosky (University of Colorado), Joel Levine (LaRC), Rob Manning (JPL),Kurt Marti (University of California, San Diego), Nick Tosca (Harvard University);

MEPAG Goal III. Bruce Banerdt (JPL), Nadine Barlow (Northern Arizona University), Steve Clifford (LPI), Jack Connerney (GSFC), BobGrimm (SwRI), Joe Kirschvink (Caltech), Laurie Leshin (GSFC), Horton Newsom (University of New Mexico), Ben Weiss (MIT);

MEPAG Goal IV. David McKay (JSC), Carl Allen (JSC), Brad Jolliff (Washington University), Paul Carpenter (Washington University),Dean Eppler (JSC), John James (JSC), Jeff Jones (JSC), Russ Kerschman (NASA/ARC), and Phil Metzger (KSC).

MEPAG ND-SAG490

VI-K. Contamination control 522VI-L. Documented sample orientation 523VI-M. Program context, and planning for the first MSR 523

VII. Summary of Findings and Recommended Follow-Up Studies 525VIII. Acknowledgments 525Appendix I (ND-SAG Charter) 526Appendix II 527Appendix III 527Appendix IV 527Appendix V 527References 529

List of Tables

Table 1 Scientific Objectives, 2003/2005 MSR, 2009 MSL, and 2013 ExoMars (Order Listed as in the Originals) 495Table 2 Planning Aspects Related to a Returned Gas Sample 506Table 3 Summary of Sample Types Needed to Achieve Proposed Scientific Objectives 508Table 4 Subdivision History of Martian Meteorite QUE 94201 510Table 5 Generic Plan for Mass Allocation of Individual Rock Samples 511Table 6 Summary of Number, Type, and Mass of Returned Samples 513Table 7 Rover-Based Measurements to Guide Sample Selection 516Table 8 Science Priorities Related to the Acquisition System for Different Sample Types 517Table 9 Effect of Maximum Sample Temperature on the Ability to Achieve the Candidate Scientific Objectives 518Table 10 Relationship of the MSL Cache to Planning for MSR 520Table 11 Science Priority of Attributes of the First MSR 524

Acronym Glossary

AMS Accelerator Mass SpectrometerAPXS Alpha Proton X-ray SpectrometerARC Ames Research CenterCOSPAR Committee on Space ResearchEDL Entry, Descent, and Landing, a critical phase for martian landersEDX Energy Dispersive AnalysisEMPA Electron Microprobe AnalysisExoMars A rover mission to Mars planned by the European Space AgencyFTIR Fourier Transform InfraredGC Gas ChromatographGCR Galactic Cosmic RaysGSFC Goddard Space Flight CenterICP Inductively Coupled PlasmaIMEWG International Mars Exploration Working GroupIMU Inertial Measurement UnitINAA Instrumental Neutron Activation AnalysisJPL Jet Propulsion LaboratoryJSC Johnson Space CenterKSC Kennedy Space CenterLC Liquid ChromatographyLaRC Langley Research CenterLD-BH Life-detection and Biohazard Testing; used in the context of the test protocolLPI Lunar and Planetary InstituteMAV Mars Ascent Vehicle, the rocket that would lift the samples off the martian surfaceMEP Mars Exploration ProgramMEPAG Mars Exploration Program Analysis GroupMER Mars Exploration Rover, a NASA mission launched in 2003MEX Mars Express, a 2003 mission of the European Space AgencyMI Microscopic Imager, an instrument on the 2003 MER missionMOD Mars Organic DetectorMOMA Mars Organic Molecule Analyzer, an instrument proposed for the 2013 ExoMars missionMRO Mars Reconnaissance Orbiter, a 2005 mission of NASAMS Mass SpectrometerMSL Mars Science Laboratory, a NASA mission to Mars scheduled to launch in 2009MSR Mars Sample Return

I. Executive Summary

THE RETURN OF MARTIAN SAMPLES TO EARTH has long beenrecognized as an essential component of a cycle of ex-ploration that begins with orbital reconnaissance and in situsurface investigations. Major questions about life, climate,and geology require answers from state-of-the-art laborato-ries on Earth. Spacecraft instrumentation cannot performcritical measurements such as precise radiometric age dat-ing, sophisticated stable isotopic analyses, and definitive life-detection assays. Returned sample studies could respondradically to unexpected findings, and returned materialscould be archived for study by future investigators with evenmore capable laboratories. Unlike martian meteorites, re-turned samples could be acquired with known context fromselected sites on Mars according to the prioritized explo-ration goals and objectives.

The ND-MSR-SAG formulated the following 11 high-levelscientific objectives that indicate how a balanced program ofongoing MSR missions could help to achieve the objectivesand investigations described by MEPAG (2006).

(1) Determine the chemical, mineralogical, and isotopiccomposition of the crustal reservoirs of carbon, nitro-gen, sulfur, and other elements with which they haveinteracted and characterize carbon-, nitrogen-, and sul-fur-bearing phases down to submicron spatial scales inorder to document processes that could sustain habit-able environments on Mars both today and in the past.

(2) Assess the evidence for prebiotic processes, past life, andextant life on Mars by characterizing the signatures ofthese phenomena in the form of structure/morphology,biominerals, organic molecular and isotopic composi-tions, and other evidence within their geologic contexts.

(3) Interpret the conditions of martian water-rock interac-tions through the study of their mineral products.

(4) Constrain the absolute ages of major martian crustal ge-ologic processes, including sedimentation, diagenesis,volcanism/plutonism, regolith formation, hydrother-mal alteration, weathering, and cratering.

(5) Understand paleoenvironments and the history of near-surface water on Mars by characterizing the clastic andchemical components, depositional processes, and post-depositional histories of sedimentary sequences.

(6) Constrain the mechanism and timing of planetary ac-cretion, differentiation, and the subsequent evolution ofthe martian crust, mantle, and core.

(7) Determine how the martian regolith was formed andmodified and how and why it differs from place toplace.

(8) Characterize the risks to future human explorers in theareas of biohazards, material toxicity, and dust/granu-lar materials and contribute to the assessment of po-tential in situ resources to aid in establishing a humanpresence on Mars.

(9) For the present-day martian surface and accessible shal-low subsurface environments, determine the preserva-tion potential for the chemical signatures of extant lifeand prebiotic chemistry by evaluating the state of oxi-dation as a function of depth, permeability, and otherfactors.

(10) Interpret the initial composition of the martian atmo-sphere, the rates and processes of atmospheric loss/gainover geologic time, and the rates and processes of atmo-spheric exchange with surface condensed species.

(11) For martian climate-modulated polar deposits, deter-mine their age, geochemistry, conditions of formation,and evolution through the detailed examination of the

SCIENCE PRIORITIES FOR MARS SAMPLE RETURN 491

ND-MSR SAG Next Decade Mars Sample Return Science Analysis Group (also abbreviated as ND-SAG)PCR Polymerase Chain ReactionPE Preliminary ExaminationPI Principal InvestigatorPLD Polar Layered DepositsRAT Rock Abrasion Tool, an instrument on the MER missionSAM Surface Analysis at Mars, an instrument on the 2009 MSL missionSEM Scanning Electron MicroscopeSIMS Secondary Ion Mass SpectrometerSMD Science Mission DirectorateSRF Sample-Receiving FacilitySSG Science Steering Group, a MEPAG subcommitteeSTEM Scanning Transmission Electron MicroscopeSwRI Southwest Research InstituteTBR To Be ReviewedTEM Transmission Electron MicroscopeTIMS Thermal Ionization Mass SpectrometerTOF-SIMS Time-of-Flight Secondary Ion Mass SpectrometerUSGS US Geological SurveyVNIR Visible/Near InfraredVS Vertical SurveysXANES X-ray Absorption Near Edge StructureXRD X-Ray Diffraction, a generic method for determining mineralogyXRF X-Ray Fluorescence, a generic method for determining sample chemistry

composition of water, CO2, and dust constituents, aswell as isotopic ratios and detailed stratigraphy of theupper layers of the surface.

MSR would attain its greatest value if samples are col-lected as sample suites that represent the diversity of theproducts of various planetary processes. Sedimentary mate-rials likely contain complex mixtures of chemical precipi-tates, volcaniclastics, impact glass, igneous rock fragments,and phyllosilicates. Aqueous sedimentary deposits are im-portant for performing measurements of life detection, ob-servations of critical mineralogy and geochemical patterns,and trapped gases. On Earth, hydrothermally altered rockscan preserve a record of hydrothermal systems that providedwater, nutrients, and chemical energy necessary to sustainmicroorganisms. They also might have preserved fossils intheir mineral deposits. Hydrothermal processes alter themineralogy of crustal rocks and inject CO2 and reduced gasesinto the atmosphere. Chemical alteration that occurs at near-surface ambient conditions (typically � �20°C) creates low-temperature altered rocks and includes, among other things,aqueous weathering and various nonaqueous oxidation re-actions. Understanding the conditions under which alter-ation proceeds at low temperatures would provide impor-tant insight into the near-surface hydrological cycle,including fluid/rock ratios, fluid compositions (chemicaland isotopic, as well as redox conditions), and mass fluxesof volatile compounds. Igneous rocks are expected to be pri-marily lavas and shallow intrusive rocks of basaltic compo-sition. They are critical for investigations of the geologic evo-lution of the martian surface and interior because theirgeochemical and isotopic compositions constrain both thecomposition of mantle sources and the processes that af-fected magmas during generation, ascent, and emplacement.Regolith samples (unconsolidated surface materials) recordinteractions between crust and atmosphere, the nature ofrock fragments, fine particles that have been moved over thesurface, exchange of H2O and CO2 between near-surfacesolid materials and the atmosphere, and processes that in-volve fluids and sublimation. Regolith studies would helpfacilitate future human exploration by assessing toxicity andpotential resources. Polar ices would constrain present andpast climatic conditions and help elucidate water cycling.Surface ice samples from the Polar Layered Deposits (PLD)or seasonal frost deposits would help to quantify surface/atmosphere interactions. Short cores could help to resolverecent climate variability. Atmospheric gas samples wouldconstrain the composition of the atmosphere and processesthat influenced its origin and evolution. Trace organic gases(e.g., methane and ethane) could be analyzed for abundances,distribution, and relationships to a potential martian bios-phere. Returned atmospheric samples that contain Ne, Kr,CO2, CH4, and C2H6 would confer major scientific benefits.Chemical and mineralogical analyses of martian dust wouldhelp to elucidate the weathering and alteration history ofMars. Given the global homogeneity of martian dust, a sin-gle sample is likely to be representative of the planet. Adepth-resolved suite of samples should be obtained fromdepths that range from cm to several m within regolith orfrom rock outcrop to investigate trends in the abundance ofoxidants (e.g., OH, HO2, H2O2, and peroxy radicals), the ef-fects of radiation, and the preservation of organic matter.

Other sample suites include impact breccias that might sam-ple rock types that are otherwise not available locally, tephraconsisting of fine-grained regolith material or layers andbeds possibly delivered from beyond the landing site, andmeteorites whose alteration history could provide insightsinto martian climatic history.

The following factors would affect our ability to achieveMSR’s scientific objectives:

(1) Sample size. A full program of scientific investigationswould likely require samples of �8 g for bedrock, looserocks, and finer-grained regolith. To support requiredbiohazard testing, each sample requires an additional 2g, leading to an optimal size of 10 g. Textural studies ofsome rock types might require one or more larger sam-ples of �20 g. Material should remain to be archivedfor future investigations.

(2) Number of samples. Studies of differences between samplescould provide more information than detailed studies ofa single sample. The number of samples needed to ad-dress MSR scientific objectives effectively is 35 (28 rock, 4regolith, 1 dust, 2 gas). If the MSR mission recovers theMSL cache, it should also collect 26 additional samples(20 rock, 3 regolith, 1 dust and 2 atmospheric gas). Thetotal mass of these samples is expected to be about 345 g(or 380 g with the MSL cache). The total returned masswith sample packaging would be about 700 g.

(3) Sample encapsulation. To retain scientific value, returnedsamples must not commingle, each sample must belinked uniquely to its documented field context, androcks should be protected against fragmentation duringtransport. A smaller number or mass of carefully man-aged samples is far more valuable than a larger num-ber or mass of poorly managed samples. The encapsu-lation of at least some samples must retain any releasedvolatile components.

(4) Diversity of the returned collection. The diversity of re-turned samples must be commensurate with the diver-sity of rocks and regolith encountered. This guidelinesubstantially influences landing site selection and roveroperation protocols. It is scientifically acceptable forMSR to visit only a single site, but visiting 2 indepen-dent landing sites would be much more valuable.

(5) In situ measurements for sample selection and documenta-tion of field context. Relatively few samples could be re-turned from the vast array of materials the MSR roverwould encounter; thus we must be able to choosewisely. At least 3 kinds of in situ observations areneeded (color imaging, microscopic imaging, and min-eralogy measurement) and possibly as many as 5 (alsoelemental analysis and reduced-carbon analysis). Nosignificant difference exists in the observations neededfor sample selection vs. sample documentation. Revis-iting a previously occupied site might result in a reduc-tion in the number of instruments.

(6) Surface operations. To collect the samples required byMSR objectives, the lander must have significant sur-face mobility and the capability to assess and samplethe full diversity of materials. Depending on the geol-ogy of the site, at least 6–12 months of surface opera-tion would be required to explore a site and assess andcollect a set of samples.

MEPAG ND-SAG492

(7) Sample acquisition system. This system must sampleweathered exteriors and unweathered interiors of rocks,as well as continuous stratigraphic sequences of out-crops that might vary in their hardness. Further, the sys-tem must have the capability to relate the orientation ofsample structures and textures to those in outcrop sur-faces, bedding planes, stratigraphic sequences, and re-gional-scale structures, and maintain the structural in-tegrity of samples. A mini-corer and a scoop are themost important collection tools. A gas compressor anda drill have lower priority but are needed for certainsamples.

(8) Sample temperature. Some key species (e.g., organics, sul-fates, chlorides, clays, ice, and liquid water) are sensi-tive to temperatures above surface temperatures. Ob-jectives could most confidently be met if samples arekept below �20°C and with less confidence if they arebelow �20°C. Significant loss, particularly to biologicalstudies, occurs if samples reach �50°C for 3 hours. Tem-perature monitoring during return would allow anychanges to be evaluated.

(9) Planning considerations involving the MSL/ExoMars caches.Retrieving the MSL or ExoMars cache might alter otheraspects of the MSR mission. However, given the limi-tations of the MSL cache, differences in planetary pro-tection requirements for MSL and MSR, the possibilitythat the cache might not be retrievable, and the poten-tial for MSR to make its own discoveries, the MSR rovershould be able to characterize and collect at least someof the returned samples.

(10) Planetary protection. A scientifically compelling firstMSR mission does not require the capability to accessand sample a special region, defined as a region withinwhich terrestrial organisms may propagate. UnlessMSR could land poleward of 30° latitude, access roughterrain, or achieve significant subsurface penetration(�5 m), MSR is unlikely to be able to use incrementalspecial regions capabilities. Planetary protection drafttest protocols should be updated to incorporate ad-vances in biohazard analytical methods. Statistical prin-ciples governing mass requirements for subsamplingreturned samples for these analyses should be re-assessed.

(11) Contamination control. Inorganic and organic contami-nation must be minimized in order to achieve MSR sci-entific objectives. A study is needed to specify samplecleanliness thresholds that must be attained duringsample acquisition and processing.

II. Introduction

Since the dawn of the modern era of Mars exploration, thereturn of martian samples to Earth has been recognized asan essential component of a cycle of exploration that beganwith orbital reconnaissance and in situ surface investigations(see, for example, the discussion of sample return in 3decades of reports by the National Research Council (e.g.,National Research Council, 1978, 1990a, 1990b, 1994, 1996,2001, 2007). Global reconnaissance and surface observationshave “followed the water” and revealed a geologically di-verse martian crust that could have sustained near-surfacehabitable environments in the distant past. However, major

questions about life, climate, and geology remain; and manyof these require answers that only Earth-based state-of-the-art analyses of samples could provide. This stems from thefact that flight instruments cannot match the adaptability, ar-ray of sample-preparation procedures, and micro-analyticalcapability of Earth-based laboratories (Gooding et al., 1989).For example, analyses conducted at the submicron scale werecrucial for investigating the ALH84001 meteorite, and theywould be essential for interpreting the returned samples.Furthermore, spacecraft instrumentation simply cannot per-form certain critical measurements, such as precise radio-metric age dating, sophisticated stable isotopic analyses, andcomprehensive life-detection experiments. If returned sam-ples yield unexpected findings, subsequent investigationscould be adapted accordingly. Moreover, portions of re-turned samples could be archived for study by future gen-erations of investigators using ever-more-powerful instru-mentation.

Some samples from Mars are available for research onEarth in the form of the martian meteorites. The martian me-teorites, while indeed valuable, provide a limited view ofmartian geologic processes. These samples are all igneous innature and minimally altered; thus they do not record thehistory of low-temperature water-based processes. Thesesamples certainly do not represent the most promising hab-itable environments (Gooding et al., 1989), and it is possiblethat the most extensively water-altered materials might betoo fragile to survive an interplanetary journey. Most mete-orites have young crystallization ages less than 1.3 billionyears, which indicates that they represent only the most re-cent igneous activity on Mars (Borg and Drake, 2005). Theirgeochemical characteristics suggest that they are closely re-lated to one another and are consequently not representativeof all the lithologic and geochemical diversity that is likelyto be present in an igneous martian rock suite (Borg andDraper, 2003; Borg et al., 2003; Symes et al., 2008). Becausethe meteorites arrived by natural processes and lack geologiccontext, it is extremely difficult to extrapolate the resultsfrom geologic studies of these samples to rocks observedfrom space or on the martian surface by landed spacecraft.In contrast, returned samples could be obtained from siteswithin a known geologic context and selected in order toachieve the goals and objectives of the Mars exploration com-munity. Nevertheless, sample-return missions must sur-mount key challenges, such as engineering complexity, cost,and planetary protection concerns, before their enormouspotential could be recognized. This document is intended todefine this critical step forward toward realizing the enor-mous potential of Mars sample return.

On July 10, 2007, Dr. Alan Stern, Associate Administratorfor the Science Mission Directorate (SMD), described to theparticipants of the 7th International Conference on Mars hisvision of achieving Mars Sample Return (MSR) no later thanthe 2020 launch opportunity. He requested that the financialattributes, scientific options/issues/concerns, and technol-ogy development planning/budgeting details of this visionbe analyzed over the next year. The Mars Exploration Pro-gram Analysis Group (MEPAG) is contributing to this effortby preparing this analysis of the science components of MSRand its programmatic context. To this end, MEPAG charteredthe Next Decade MSR Science Analysis Group (ND-MSR-SAG) to complete 4 specific tasks:

SCIENCE PRIORITIES FOR MARS SAMPLE RETURN 493

(1) Analyze what critical Mars science could be accom-plished in conjunction with, and complementary to, anext decade MSR mission.

(2) Evaluate the science priorities associated with guidingthe makeup of the sample collection to be returned byMSR.

(3) Determine the dependencies of mobility and surface life-time of MSR on the scientific objectives, sample acquisi-tion capability, diagnostic instrument complement, andnumber and type of samples.

(4) Support MSR science planning as requested by the In-ternational Mars Exploration Working Group (IMEWG)MSR study. The charter is presented in Appendix I.

The return of any reasonable sample mass from Marswould significantly increase our understanding of atmo-spheric, biologic, and geologic processes occurring there, aswell as permit evaluation of the hazards to humans on thesurface. This is largely independent of how the samples areselected, collected, and packaged for return, and stems fromthe fact that there are no analogous samples on Earth. Thus,a mission architecture in which a limited number of surfacesamples are collected in a minimum amount of geologic con-text has been recommended in the past and has huge scien-tific merit (e.g., MacPherson et al., 2005). It is also importantto realize that a significantly greater scientific yield wouldresult from samples that are more carefully selected. Ana-lytical results from samples that are screened, placed in de-tailed geologic context, collected from numerous locationsand environments, and packaged and transported underconditions that more closely approximate those encounteredon the martian surface would dramatically clarify the pic-ture of Mars derived from the mission, as well as allow an-alytical results to be more rigorously extrapolated to theplanet as a whole. As a consequence of these facts, this doc-ument outlines a sampling strategy that is necessary to max-imize scientific yield. The inability to complete all the surfaceoperations associated with this sampling strategy by nomeans negates the usefulness of these samples. Rather, it re-sults in a proportional loss of scientific yield of the mission.Thus, this study is expected to constitute input to a Marsprogram architecture trade analysis between scientific yieldand cost.

III. Evaluation Process

Prior to beginning this study, the ND-SAG was briefed onthe conclusions of the NASA Mars Sample Return ScienceSteering Group II (MacPherson et al., 2005; Appendix III) andthe National Research Council Committee on an Astrobiol-ogy Strategy for the Exploration of Mars. These reports document the importance of sample return in a completestrategy for the exploration of Mars, and many of their con-clusions are reiterated here. However, the current analysishas benefited from discoveries made in the interval sincethese reports were written, such as phyllosilicates, silica, andthe distribution and context of polyhydrated sulfates on thesurface of Mars. It is expected that some of the conclusionsof this report will be further elucidated and strengthened asresults from Phoenix, MSL, and ExoMars become available.This may be particularly true of the results from analyses oforganic matter and ices.

Assumptions used in this study are:

(1) The sample return mission would begin in either 2018 or2020.

(2) MSL will launch in 2009 and will prepare a rudimentarycache of samples that would be recoverable by the MSRmission. ExoMars would carry a similar cache.

(3) The functionality of sample acquisition associated withMSR would be independent of MSL. This functionalitymay either be landed at the same time as the sample re-turn element of MSR, or it may be separated into a pre-cursor mission.

(4) The Mars Exploration Program would maintain a stableprogram budget of about $625 million per year thatgrows at 2% per year.

To complete these tasks and link strongly the report of theND-SAG to the MEPAG Goals Document, the ND-SAG wasdivided into 4 subteams that correspond to each of the 4 mainMEPAG goals. The goals, as outlined in the Goals Document,are: determine whether life ever arose on Mars, understandthe processes and history of climate on Mars, determine theevolution of the surface and interior of Mars, and preparefor human exploration. Each group examined the individualinvestigations outlined in the MEPAG Goals Document andconsidered the following:

• Whether sample return would facilitate the investigation.• The type, mass, number, and diversity of samples that

would be required to complete the investigation.• The physical condition of the samples (rock, pulverized

rock, etc.).• The vulnerability of specific sample types to degradation

effects during sample collection, encapsulation, and trans-port, as well as the impact of this degradation on indi-vidual investigations.

• The measurements required at the time of sample collec-tion in order to select appropriate samples and place themin the necessary geologic context.

• The mobility necessary to obtain required samples.• The packaging and handling priorities necessary to pre-

serve the characteristics of interest in the samples.

The results of this analysis are presented in detail in Ap-pendix II. Below, we summarize the consensus of the ND-SAG that was derived from this analysis.

IV. Scientific Objectives of MSR

IV-A. History, current context of MSR’s scientificobjectives

The 2003/2005 Mars Sample Return mission (which was can-celled in 2000, prior to launch) was the most recent effort thatformulated scientific objectives for MSR. The way this missionchose to frame its scientific objectives is shown in Table 1. Since2000, there have been numerous scientific advances that havegreatly increased our understanding of the Red Planet. It is crit-ical to take these into consideration in setting the new scientificobjectives for MSR. In particular, it is important to incorporateactual or anticipated results from the following:

Recent and ongoing flight missions. Since the last MSRanalysis in 2000, the Mars Global Surveyor (1999–2006),

MEPAG ND-SAG494

TA

BL

E1.

SCIE

NT

IFIC

OB

JEC

TIV

ES,

200

3/20

05 M

SR, 2

009

MSL

, AN

D20

13 E

XO

MA

RS

(OR

DE

RL

IST

ED

AS

INT

HE

OR

IGIN

AL

S)

MSR

(20

03/2

005)

MSL

(20

09)

Exo

Mar

s (2

013)

1Fu

rthe

r ou

r un

der

stan

din

g of

the

1C

hara

cter

izat

ion

of g

eolo

gic

feat

ures

, con

trib

utin

g to

1T

o se

arch

for

sig

ns o

f pa

st a

nd p

rese

ntpo

tent

ial

and

pos

sibl

e bi

olog

ical

dec

iphe

ring

geo

logi

c hi

stor

y an

d t

he p

roce

sses

tha

tlif

e on

Mar

shi

stor

y of

Mar

sha

ve m

odif

ied

roc

ks a

nd r

egol

ith,

inc

lud

ing

the

role

of

wat

er2

Sear

ch f

or i

ndic

ator

s of

pas

t or

2D

eter

min

atio

n of

the

min

eral

ogy

and

che

mic

al2

To

char

acte

rize

the

wat

er/

geoc

hem

ical

pres

ent

life

on t

he m

arti

an s

urfa

ceco

mpo

siti

on (

incl

udin

g su

ch a

n in

vent

ory

of e

lem

ents

suc

hd

istr

ibut

ion

as a

fun

ctio

n of

dep

th i

nas

C, H

, N, O

, P, S

, etc

. kno

wn

to b

e bu

ildin

g bl

ocks

the

shal

low

sub

surf

ace

for

life)

of

surf

ace

and

nea

r-su

rfac

e m

ater

ials

3Im

prov

e ou

r un

der

stan

din

g of

3D

eter

min

atio

n of

ene

rgy

sour

ces

that

cou

ld b

e us

ed t

o3

To

stud

y th

e su

rfac

e en

viro

nmen

t an

dm

arti

an c

limat

ic e

volu

tion

and

sust

ain

biol

ogic

al p

roce

sses

iden

tify

haz

ard

s to

fut

ure

hum

anpl

anet

ary

hist

ory

mis

sion

s4

Impr

ove

our

und

erst

and

ing

of4

Cha

ract

eriz

atio

n of

org

anic

com

poun

ds

and

pot

enti

al4

To

inve

stig

ate

the

plan

et’s

sub

surf

ace

cons

trai

nts

on t

he a

mou

nt a

ndbi

omar

kers

in

repr

esen

tati

ve r

egol

ith,

roc

ks, a

nd i

ces

and

dee

p in

teri

or t

o be

tter

und

erst

and

hist

ory

of w

ater

on

and

wit

hin

Mar

sth

e ev

olut

ion

and

hab

itab

ility

of

Mar

s5

Acq

uire

dat

a to

id

enti

fy a

reas

of

5D

eter

min

atio

n of

the

sta

ble

isot

opic

and

nob

le g

aspo

ssib

le i

nter

est

for

futu

re s

cien

tifi

cco

mpo

siti

on o

f th

e pr

esen

t-d

ay b

ulk

atm

osph

ere

expl

orat

ion

6D

eter

min

e th

e na

ture

of

loca

l su

rfac

e6

Iden

tifi

cati

on o

f po

tent

ial

bios

igna

ture

s (c

hem

ical

,ge

olog

ic p

roce

sses

fro

m s

urfa

cete

xtur

al, i

soto

pic)

in

rock

s an

d r

egol

ith

mor

phol

ogy

and

che

mis

try

7D

eter

min

e th

e sp

atia

l d

istr

ibut

ion

7C

hara

cter

izat

ion

of t

he b

road

spe

ctru

m o

f su

rfac

ean

d c

ompo

siti

on o

f m

iner

als,

roc

ks,

rad

iati

on, i

nclu

din

g ga

lact

ic c

osm

ic r

adia

tion

, sol

aran

d s

oils

sur

roun

din

g th

e la

ndin

gpr

oton

eve

nts,

and

sec

ond

ary

neut

rons

site

s8

Cha

ract

eriz

atio

n of

the

loc

al e

nvir

onm

ent,

incl

udin

gba

sic

met

eoro

logy

, the

sta

te a

nd c

yclin

g of

wat

er a

ndC

O2

and

the

nea

r-su

rfac

e d

istr

ibut

ion

of h

ydro

gen

Sour

ces

of in

form

atio

n: M

SR: O

’Nei

l and

Caz

aux

(200

0); M

SL: h

ttp:

//m

epag

.jpl.n

asa.

gov/

MSL

_Sci

ence

_Obj

ecti

ves.

htm

l (as

of J

anua

ry 7

, 200

8); E

xoM

ars:

Vag

o an

d K

min

ek (2

007)

.

Mars Odyssey (2002–present), Mars Exploration Rovers(2004–present), Mars Express (mapping from 2004–present),and the Mars Reconnaissance Orbiter (mapping from2006–present) have made important discoveries. These in-vestigations have greatly improved our understanding ofMars and have resulted in progressive refinement of keymartian scientific objectives, as documented by the evolutionof the MEPAG Goals Document (MEPAG, 2001, 2004, 2005,2006).

Future (but pre-MSR) flight missions. Two major missionsto the martian surface are scheduled during the next 6 years—the Mars Science Laboratory (MSL, scheduled for launch in2009) and ExoMars (scheduled for launch in 2013). Both mis-sions will analyze rock samples on the surface of Mars usingin situ methods. It is therefore necessary to consider the scien-tific objectives of these missions when planning the objectivesof the first MSR mission and to build upon their expected ac-complishments. The scientific objectives of the MSL and Exo-Mars missions, as of 2007, are listed in Table 1.

Meteorite studies. More than 35 martian meteorites havebeen found in Antarctica and desert environments by mete-orite recovery programs, including private and government-sponsored efforts. The number of recovered meteorites continually increases. As a consequence, MSR scientific ob-jectives and sample selection strategy must respond to sci-entific advances derived from meteorite studies and alsocomplement the existing meteorite collections.

IV-B. Possible scientific objectives for a next decade MSR

To translate the general statements about the possiblevalue of MSR into specifics (Appendix II), the ND-SAG an-alyzed how returned samples might contribute to each of thescientific objectives and investigations described by MEPAG(2006). The investigations listed in MEPAG (2006) do nothave equal science priority, nor do they benefit equally fromreturned sample analyses. By considering the most impor-tant potential uses of returned samples, the ND-SAG has for-mulated 11 relatively high-level scientific objectives for MSR.We note, however, that no single landing site could addressall these objectives. Those objectives that any single MSRmission could achieve would reflect the capabilities of its ar-chitecture/hardware and the geologic terrain and local cli-mate of the site. Even though all these objectives could notbe achieved on the first MSR mission, it is ND-SAG’s hopethat by making this analysis as complete as possible, it willset the scene for future MSR missions beyond the first one.

Prioritization of the scientific objectives. The ND-SAGteam considered the relative priority of the possible objec-tives listed below, using the following prioritization criteria:

(1) The investigation priority in the Goals Document(MEPAG, 2006). The analysis in Appendix II finds thatreturned samples could significantly advance 34 of theinvestigations identified by MEPAG (2006), and each ofthese investigations has been assigned a priority byMEPAG. The way in which these 34 investigations areconsolidated into the 11 objective statements below isshown graphically in Appendix IV.

(2) The impact of MSR on investigation(s) associated withthese objectives. For 13 of the 34 investigations, MSRwould not only be expected to advance them more sub-stantially than for the others; in some cases, MSR is es-sential (shown by the color coding in Appendix IV).

However, the achievable degree of progress toward thesescientific questions would also depend on the choice of land-ing site (and the kinds of samples that are available to be col-lected there), the capability of the engineering system (e.g.,the number and quality of samples), the degree of complex-ity of the geologic process under study (and how many sam-ples it might take to evaluate it), and other factors. For ex-ample, Objective 5 involves processes that are very complex,and a quantum jump in our understanding may be difficultwith only a few samples. However, the objective is clearlyimportant, and we should let it help guide the engineering.For these reasons, the ND-SAG team felt it appropriate tolist the scientific objectives below in only 2 general prioritygroups: the first 5 are considered high priority, and the last6 are considered medium priority. These priorities willclearly need to be reconsidered as the specifics of MSR arerefined.

(1) Determine the chemical, mineralogical, and iso-topic composition of the crustal reservoirs of carbon,nitrogen, sulfur, and other elements with which theyhave interacted and characterize carbon-, nitrogen-,and sulfur-bearing phases down to submicron spatialscales in order to document processes that could sus-tain habitable environments on Mars, both today andin the past.

Discussion. A critical assessment of the habitability of past andpresent martian environments must determine how the elementalbuilding blocks of life have interacted with crustal and atmosphericprocesses (Des Marais et al., 2003). On Earth, such interactions havedetermined the bioavailability of these elements, the potential sourcesof biochemical energy, and the chemistry of aqueous environments(e.g., Konhauser, 2007). Earth-based investigations of martian me-teoritic minerals, textures, and chemical composition at the submi-cron scale have yielded discoveries of their igneous volatiles, impact-related alteration, carbonates, organic carbon, atmosphericcomposition, and the processes that shaped them. The search for ex-tant life requires exploration of special regions (sites where life mightbe able to propagate) and thereby invokes stringent planetary pro-tection protocols. These protocols are less stringent at sites other thanspecial regions where the search for past life would target fossil biosig-natures preserved in rocks. This objective is an extension of MSL Ob-jectives 1 through 4 (Table 1), ExoMars Objectives 2 and 4 (Table1), and MEPAG Objective I-A, which collectively address the habit-ability potential of martian environments.

(2) Assess the evidence for prebiotic processes, pastlife, and extant life on Mars by characterizing the sig-natures of these phenomena in the form of struc-ture/morphology, biominerals, organic molecular andisotopic compositions, and other evidence within theirgeologic contexts.

Discussion. The MER mission demonstrated that habitableenvironments existed on Mars in the past and that their geologic

MEPAG ND-SAG496

deposits are accessible at the surface (Squyres and Knoll, 2005a,2005b; Des Marais et al., 2007). The Mars Express OrbiterOMEGA IR spectrometer mapped aqueous minerals that formedduring the Noachian (Bibring et al., 2005; Poulet et al., 2005).The upcoming MSL and ExoMars missions will be able to provideinformation about the habitability (past or present) of their specificlanding sites at even greater detail. Although ExoMars is designedto search for traces of past and present life (it should also be ableto detect prebiotic organic materials), experience with martian me-teorites and, more especially, microfossil-containing rocks fromearly Earth, has shown that identifying traces of life reliably is ex-traordinarily difficult because (1) microfossils are often very smallin size and (2) the quantities of organic carbon in the rocks thatare identifiable as biogenic or abiogenic are often very low (West-all and Southam, 2006). The reliable identification of mineral andchemical biosignatures typically requires some particular combi-nation of sophisticated high-resolution analytical microscopes,mass spectrometers, and other advanced instrumentation. The par-ticular combination of instruments that are most appropriate andeffective for a given sample is often determined by the initial analy-ses. Accordingly, sample measurements must be conducted onEarth because they require adaptability in the selection of advancedinstrumentation. Note that the specifics of how this objective ispursued would be highly dependent on landing site selection. Thesearch for extant life would require that the rover meet planetaryprotection requirements for visiting a “special region.” The local-ities that are judged to be most prospective for evaluating prebi-otic chemistry and fossil life might not be the most favorable forextant life. However, all returned samples would assuredly be eval-uated for evidence of extant life, in part to fulfill planetary pro-tection requirements, whether or not the samples were targeted forthis purpose. This objective is an extension of MSL Objective 6(Table 1), ExoMars Objective 1 (Table 1), and MEPAG ObjectivesI-A, I-B, and I-C, which address habitability, prebiotic chemistry,and biosignatures.

(3) Interpret the conditions of martian water-rock in-teractions through the study of their mineral products.

Discussion. Both igneous and sedimentary rocks are suscep-tible to water-rock interactions that range from low-temperatureweathering through hydrothermal interactions. These processescould operate from the surface to great depths within the martiancrust. Rocks and minerals affected by such processes are signifi-cant repositories of volatile light elements in the martian crust,and they have also recorded evidence of climate and crustal pro-cesses, both past and present. The compositions and textures ofrock and mineral assemblages frequently reveal the water-to-rockratios, fluid compositions, and environmental conditions that cre-ated those assemblages [also discussed by MacPherson et al.(2001)]. A significant fraction of the key diagnostic informationexists as rock textures, crystals, and compositional heterogeneitiesat submicrometer to nanometer spatial scales. Textural relation-ships between mineral phases could help to determine the order ofprocesses that have affected the rocks. This is key to determining,for example, whether a rock is of primary aqueous origin or, al-ternatively, was affected by water at some later time in its history.Accordingly, state-of-the art Earth-based laboratories are requiredto read the record of water-rock interactions and infer their sig-nificance for the geologic and climatic history of Mars. This ob-jective is an extension of the discoveries of MRO, MEX, and MERthat there is an extensive history of ancient interaction between

water and the martian crust. Understanding these interactionsover a broad range of spatial scales is critical for interpreting thehydrologic record and records of thermal and chemical environ-ments. This objective is an extension of MSL Objectives 1, 2, and8 (Table 1), ExoMars Objectives 2 and 4 (Table 1), and MEPAGObjectives I-A, II-A, III-A, and IV-A.

(4) Constrain the absolute ages of major martiancrustal geologic processes, including sedimentation,diagenesis, volcanism/plutonism, regolith formation,hydrothermal alteration, weathering, and cratering.

Discussion. Constraining the absolute ages of martian rock-forming processes is an essential part of understanding Mars as asystem. There are 2 aspects to this objective. First, dating indi-vidual flow units with known crater densities would provide a cal-ibration of martian cratering rates. This is critical for the inter-pretation of orbital data because crater chronology is the primarymethod for interpreting both relative and absolute ages of geologicunits from orbit, and the method can be applied on a planetaryscale. The scientific community has strongly advocated for the cal-ibration of the crater chronology method since the inception of theMars Exploration Program (MEPAG Investigation III-A-3). Sec-ond, we need to understand the timing of different geologic pro-cesses in the past as the planet has evolved in time and space. Thesuitability of the products of different geologic processes to themethods of radiometric geochronology depends on when the iso-topic systems closed. Igneous rocks are by far the most useful [seesummary in Borg and Drake (2005)]. Constraints on low-temper-ature processes, such as sedimentation, weathering, and diagene-sis, could be obtained most easily and definitively by finding sitesthat show discernable field relationships with datable igneous ma-terials. For example, by determining the ages of igneous rocks thatare interbedded with sedimentary rocks, the interval of time whenthe sediments were deposited could be constrained. In addition, theages of secondary alteration of martian meteorites have been mea-sured with some success (Borg et al., 1999; Shih et al., 1998; 2002;Swindle et al., 2000). Accordingly, chemical precipitates formedduring diagenesis, hydrothermal activity, and weathering may bedatable with Ar-Ar, Rb-Sr and Sm-Nd chronometers. However,sophisticated Earth-based laboratories would be required so as toperform these difficult measurements precisely, with multiplechronometers to provide an internal cross-check and interpret themeanings of these ages. This objective is an extension of MSL Ob-jective 1 (Table 1), ExoMars Objective 4 (Table 1), and MEPAGObjectives I-A, II-B, III-A, and III-B, and has long been consid-ered a major objective of MSR (e.g., MacPherson et al., 2001;2002).

(5) Understand paleoenvironments and the history ofnear-surface water on Mars by characterizing the clas-tic and chemical components, depositional processes,and post-depositional histories of sedimentary se-quences.

Discussion. Experience with the Mars Exploration RoversSpirit and Opportunity demonstrates that sedimentary rock se-quences, which include a broad range of clastic and chemical con-stituents, are exposed and that sedimentary structures and bed-ding are preserved on the martian surface. Discoveries by MROand Mars Express further demonstrate the great extent and geo-logic diversity of such deposits. Sedimentary rocks could retain

SCIENCE PRIORITIES FOR MARS SAMPLE RETURN 497

high-resolution records of a planet’s geologic history, and theycould also preserve fossil biosignatures. As such, sedimentary se-quences are among the targets being considered by MSL and Ex-oMars. Previous missions have also demonstrated that the sedi-mentologic and stratigraphic character of these sequences could beevaluated with great fidelity, comparable to that attained by sim-ilar studies on Earth (e.g., Squyres and Knoll, 2005a, 2005b;Squyres et al., 2007a, 2007b). The physical, chemical, and isotopiccharacteristics of such sequences would reveal the diversity of en-vironmental conditions of the martian surface and subsurface be-fore, during, and after deposition. But much of the key diagnosticinformation in these sequences occurs as textures, minerals, andpatterns of chemical composition at the submicron scale. Futurerobotic missions might include microscopic imaging spectrometersto examine these features. However, definitive observations of suchfeatures probably would also require thin-section petrography,SEM, TEM, and other sophisticated instrumentation availableonly in state-of-the-art Earth-based laboratories. This objective isan extension of MSL Objectives 1, 2, and 8 (Table 1), ExoMarsObjectives 2 and 4 (Table 1), and MEPAG Objectives I-A, II-A,III-A, and IV-A .

(6) Constrain the mechanism and timing of planetaryaccretion, differentiation, and the subsequent evolu-tion of the martian crust, mantle, and core.

Discussion. Studies of martian meteorites have provided a fas-cinating glimpse into the fundamental processes and timescales ofaccretion (e.g., Wadhwa, 2001; Borg et al., 2003; Shearer et al.,2008; Symes et al., 2008) and subsequent evolution of the crust,mantle, and core (e.g., Treiman, 1990; Shearer et al., 2008). Mar-tian meteorites also record a history of fluid alteration as shownby the presence of microscopic clay and carbonate phases (e.g.,Gooding et al., 1991; McKay et al., 1996; Bridges et al., 2001).Although the trace-element and isotopic variability of the martianmeteorite suite far exceeds that observed in equivalent suites ofbasalts from Earth and the Moon (Borg et al., 2003), the appar-ent diversity of igneous rocks identified by both orbital and sur-face missions far exceeds that of the meteorite collection. This im-plies that an extensive record of the differentiation and evolutionof Mars has been preserved in igneous lithologies that have notbeen sampled. Samples returned from well-documented martianterrains would provide a broader planetary context for the previ-ous studies of martian meteorites and also lead to significant in-sights into fundamental crustal processes beyond those revealed bythe martian meteorites. Key questions include the following: (1)When did the core, mantle, and crust first form? (2) What are thecompositions of the martian core, mantle, and crust? (3) What ad-ditional processes have modified the crust, mantle, and core; andhow have these reservoirs interacted through time? (4) What pro-cesses produced the most recent crust? (5) What is the evolution-ary history of the martian core and magnetic field? (6) How com-positionally diverse are mantle reservoirs? (6) What are the thermalhistories of the martian crust and mantle, and how have they con-strained convective processes? (7) What is the nature of fluid-basedalteration processes in the martian crust? Coordinated studies ofmartian meteorites and selected martian samples involving detailedisotopic measurements in multiple isotopic systems, studies of mi-croscopic textural features (melt inclusions, shock effects), andcomparative petrology and geochemistry are needed to answer thesequestions definitively. These data would provide the basis for modelages of differentiation that are placed in the context of Solar Sys-

tem evolution. They would also permit some of the compositionalcharacteristics of crust, mantle, and core to be determined, whichin turn would allow geologic interactions between these reservoirsto be evaluated and their thermal histories elucidated. The tremen-dous value of this approach has been validated by geochemical stud-ies on the returned lunar samples that have been more informativethan any other means in deciphering the geologic history of theMoon. This objective is an extension of MSL Objective 1 (Table1), ExoMars Objective 4 (Table 1), and MEPAG Objectives I-A,II-A, III-A, and III-B and has long been considered a major objec-tive of MSR (e.g., MacPherson et al., 2001, 2002).

(7) Determine how the martian regolith was formedand modified, and how and why it differs from placeto place.

Discussion. The martian regolith preserves a record of crustal,atmospheric and fluid processes. Regolith investigations would de-termine and characterize the important ongoing processes that haveshaped the martian crust and surface environment during its his-tory. The martian crust is a combination of broken/disaggregatedcrustal rocks, impact-generated components (Schultz and Mus-tard, 2004), volcanic ash (Wilson and Head, 2007), oxidized com-pounds, ice, aeolian deposits, and meteorites. The Viking,Pathfinder, and MER landers have also revealed diverse mineralassemblages within regolith that include hematite nodules, salt-rich duricrusts, and silica-rich deposits (e.g., Ruff et al., 2007;Wänke et al., 2001) that show local fluid-based alteration. The re-golith contains fragments of local bedrock as well as debris thatwere transported regionally or even globally. These materialswould accordingly provide local, regional, and global contexts forgeologic and geochemical studies of the returned samples. Martiansurface materials have also recorded their exposure to cosmic-rayparticles. Cosmic-ray exposure ages obtained at Apollo landingsites have helped to date lunar impact craters (e.g., Eugster, 2003).Regolith returned from Mars should provide similar informationthat could in turn be used to constrain the absolute ages of localmartian terrains. An MSR objective would be to examine returnedsamples of regolith mineral assemblages in order to determine theabundances and movement of volatile-forming elements and anyorganic compounds in near-surface environments and to determinetheir crustal inventories. The abundance of ice in the regolith variesdramatically across the martian surface. At high latitudes, waterice attains abundances of tens of weight percent below the top fewtens of cm. Inventories of water ice at near-equatorial latitudes areless understood, but ice might occur below the top few cm (Feld-man et al., 2004). The regolith is assumed to harbor a large frac-tion of the martian CO2 and H2O inventories, but their abundancehas not yet been accurately determined. This objective is an ex-tension of MSL Objectives 1, 2, 3, 4, 6, 7, and 8 (Table 1), Exo-Mars Objectives 1, 2, and 3 (Table 1), and MEPAG Objectives I-A, I-B, I-C, II-B, III-A, and IV-A.

(8) Characterize the risks to future human explorers inthe areas of biohazards, material toxicity, anddust/granular materials, and contribute to the assess-ment of potential in situ resources to aid in establish-ing a human presence on Mars.

Discussion. Returned samples could help to accomplish 4 tasksthat are required to prepare for human exploration of Mars (seeAppendix II). These tasks include: (1) Understanding the risks that

MEPAG ND-SAG498

granular materials at the martian surface present to the landedhardware (Investigation IV-A-1A), (2) Determining the risk asso-ciated with replicating biohazards (i.e., biological agents, Investi-gation IV-A-1C), (3) Evaluating possible toxic effects of martiandust on humans (Investigation IV-A-2), and (4) Expanding knowl-edge of potential in situ resources (Investigation IV-A-1D). Thehuman exploration community has consistently advocated thatthese tasks are essential for understanding the hazards and plan-ning the eventual human exploration of Mars at an acceptable levelof risk (Davis, 1998; National Research Council, 2002; Jones etal., 2004). Regarding possible martian biohazards, analyses of ro-botically returned martian samples might be required before hu-man missions could commence in order to quantify their medicalbasis and to address concerns related to planetary protection fromboth a forward- and back-contamination perspective (Warmflashet al., 2007). This objective is an extension of MSL Objective 7(Table 1), ExoMars Objective 3 (Table 1), and MEPAG ObjectiveIV-A.

(9) For the present-day martian surface and accessibleshallow subsurface environments, determine thepreservation potential for the chemical signatures ofextant life and prebiotic chemistry by evaluating thestate of oxidation as a function of depth, permeability,and other factors.

Discussion. The surface of Mars is oxidizing, but the compo-sition and properties of the responsible oxidant(s) are unknown.Characterizing the reactivity of the near surface of Mars, includ-ing atmospheric (e.g., electrical discharges) and radiation processesas well as chemical processes with depth in the regolith and withinweathered rocks, is critical with regard to investigating in greaterdetail the nature and abundance of any organic carbon on the sur-face of Mars. Understanding the oxidation chemistry and thoseprocesses that control its variations would aid in predicting sub-surface habitability, if no organics are found on the surface, andalso in understanding how such oxidants might participate in re-dox reactions that could provide energy for life. Potential mea-surements include the identification of species and concentrationsof oxidants, characterization of the processes that form and destroythem, and characterization of concentrations and fluxes of redox-sensitive gases in the lower atmosphere. Measuring the redox statesof natural materials is difficult and may require returned samples.This objective is an extension of MSL Objectives 1 and 8 (Table1), ExoMars Objective 2 (Table 1), and MEPAG Objectives I-A,III-A, and IV-A.

(10) Interpret the initial composition of the martianatmosphere, the rates and processes of atmosphericloss/gain over geologic time, and the rates and pro-cesses of atmospheric exchange with surface con-densed species.

Discussion. The modern chemistry of the martian atmospherereflects the integration of 3 major processes, each of which is of ma-jor importance to understanding Mars: (1) The initial formationof the atmosphere, (2) The various processes that have resulted inadditions or losses to the atmosphere over geologic time, and (3)The processes by which the atmosphere exchanges with variouscondensed phases in the upper crust (e.g., ice, hydrates, and car-bonates). Many different factors have affected the chemistry of themartian atmosphere; however, if the abundance and isotopic com-

position of its many chemical components could be measured withsufficient precision, definitive interpretations are possible. We havealready gathered some information about martian volatiles fromisotopic measurements by Viking and on martian meteorites(Owen et al., 1977; Bogard et al., 2001). In addition, MSL willhave the capability to measure some, but not all, of the gas speciesof interest with good precision. This leaves 2 planning scenarios:If for some reason MSL does not deliver its expected data on gaschemistry, this scientific objective would become quite importantfor MSR. However, even if MSL is perfectly successful, it will notbe able to measure all the gas species of interest at the precisionneeded, so returning an atmospheric sample could still be an im-portant scientific objective for MSR. This objective is an extensionof MSL Objective 5 (Table 1) and MEPAG Objectives I-A, II-A,II-B, and III-A.

(11) For martian climate-modulated polar deposits, de-termine their age, geochemistry, conditions of forma-tion, and evolution through the detailed examinationof the composition of water, CO2, and dust con-stituents, and determine isotopic ratios and detailedstratigraphy of the upper layers of the surface.

Discussion. The PLD represent a detailed record of recentmartian climatic history. The composition of the topmost few me-ters of ice reflect the influence of meteorology, depositional episodes,and planetary orbital/axial modulation over the timescales of or-der 105 to 106 years (Milkovich and Head, 2005). This objectiveaddresses the priorities of MEPAG Investigation II-B-5. Terres-trial ice cores have contributed fundamentally to interpretingEarth’s climatic history. Similar measurements of martian icescould be expected to reveal critical information about that planet’sclimatic history and its surface/atmosphere interactions (Petit etal., 1999; Hecht et al., 2006). The ability of ice to preserve organiccompounds (and, potentially, organic biosignatures) may help ad-dress objectives associated with habitability and prebiotic chem-istry and life (MEPAG Goal 1; Christner et al., 2001). By ex-ploring lateral and vertical stratigraphy of active ice layers andfacilitating state-of-the-art analyses of returned materials, a rover-equipped sample return mission would significantly improve ourunderstanding beyond what the Phoenix stationary lander is ex-pected to achieve at its single high-latitude site. This objective isan extension of MEPAG Objectives I-A, II-A, II-B, and III-A.

V. Samples Required to Achieve the ScientificObjectives

The MSR scientific objectives imply the return of severaltypes of martian samples. These types arise from the varietyof significant processes (e.g., igneous, sedimentary, hy-drothermal, aqueous alteration, etc.) that played key roles inthe formation of the martian crust and atmosphere. Each pro-cess creates varieties of materials that differ in their compo-sition, location, etc., and that collectively could be used tointerpret that process. Accordingly, we define a “samplesuite” as the set of samples required to determine the keyprocess(es) that formed them. On Earth, suites typically con-sist of a few to hundreds of samples, depending on the na-ture, scale, and detail of the process(es) being addressed.However, as discussed in a subsequent section, suites ofabout 5 to 8 samples are thought to represent a reasonablecompromise between scientific needs and mission con-

SCIENCE PRIORITIES FOR MARS SAMPLE RETURN 499

straints. The characteristics of each type of sample suite arepresented below.

V-A. Sedimentary materials rock suite

Sedimentary materials would be a primary sampling ob-jective for MSR. Data from surface-roving and orbiting in-struments indicate that lithified and unlithified sedimentarymaterials on Mars likely contain a complex mixture of chem-ical precipitates, volcaniclastic materials and impact glass,igneous rock fragments, and phyllosilicates (McLennan andGrotzinger, 2008). Chemical precipitates detected or ex-pected in martian materials include sulfates, chlorides, sil-ica, iron oxides, and, possibly, carbonates and borates(McLennan and Grotzinger, 2008). Sand- to silt-sized igneousrock fragments are likely to be the dominant type of silici-clastic sediment on Mars. Sediments rich in phyllosilicatesare inferred to derive from basaltic to andesitic igneous rocksthat have undergone weathering that led to the formation ofclay minerals and oxides (Poulet et al., 2005; Clark et al., 2007).Products of weathering are moved by transporting agentssuch as wind, gravity, and water to sites of deposition andaccumulation. Sedimentary materials accumulate by addi-tion of new material on the top of the sediment column,which thereby permits historical reconstruction of conditionsand events starting from the oldest at the bottom and con-tinuing to the youngest at the top of a particular depositionalsequence. However, pervasive impacts have “gardened”(stirred and disrupted) many such layered sedimentary de-posits; therefore, undisturbed sequences must be sought. Al-though hydrothermal deposits and in situ low-temperaturealteration products of igneous rocks are products of sedi-ment-forming processes, they are presented in separate sec-tions to emphasize their importance.

Chemical precipitates formed under aqueous conditionscould be used to constrain the role of water in the martiansurface environment (e.g., Clark et al., 2005; Tosca et al., 2005).Precipitates could form within the water column and settleto the sediment surface, or they could crystallize directly onthe sediment surface as a crust. Any investigation that in-volves habitability, evidence of past or present life, climateprocesses, or evolution of the martian atmosphere would beenabled by the acquisition of these rocks (Farmer and DesMarais, 1999). Some, but not all, chemical precipitates haveinterlocking crystalline textures with low permeability,which would potentially allow preservation of trapped la-bile constituents such as organic compounds and sulfides(e.g., Hardie et al., 1985). Thus, intact samples of chemicalprecipitates would be critical for unraveling the history ofaqueous processes, including those that have influenced thecycling of carbon and sulfur.

Siliciclastic sedimentary materials are moved as solid par-ticles and deposited when a transporting agent loses energy.Variation in grain size and textural structures at scales frommillimeters to meters are important indicators of deposi-tional processes and changing levels of energy in the envi-ronment (Grotzinger et al., 2005). Secondary mineralizationof sedimentary materials is likely to be minimal if porespaces are filled with dry atmospheric gases but is likely tobe substantial if pore spaces are filled with freshwater orbrine (McLennan et al., 2005). Sub-millimeter textures atgrain boundaries are indicative of processes that have mod-

ified the sedimentary deposit. Thus, individual samples ofsiliciclastic sedimentary materials would provide insightsinto transporting agents, chemical reactions, availability ofwater in surface environments, and the presence of currentsor waves. A series of samples through a sedimentary se-quence would provide critical insight into rates and magni-tudes of sedimentary processes. Certain deposits such aschemically precipitated sediments, varved sediments, ice,etc., could provide insights into climatic cycles. Siliciclasticsedimentary materials are central to investigations that in-volve past and present habitability and the evolution of themartian surface. Fine-grained siliciclastic materials rich inphyllosilicates are likely to have low permeability and thusincrease the potential for preservation of co-deposited or-ganic matter and sulfide minerals (Potter et al., 2005). Likechemical precipitates, samples of phyllosilicates that weredeposited in aqueous environments would be critical for un-raveling the carbon and sulfur cycle on Mars.

V-B. Hydrothermal rock suite

Hydrothermal deposits are relevant to the search for tracesof life on Mars for several reasons (Farmer, 1998). On Earth,such environments can sustain high rates of biological pro-ductivity (Lutz et al., 1994). The microbial life-forms that in-habit these environments benefit from various thermody-namically favorable redox reactions, such as those thatinvolve hot water and mineral surfaces. These conditions canalso facilitate the abiotic synthesis of organics from CO2 orcarbonic acid (McCollom and Shock, 1996). The kinds of molecules that are thus synthesized include monomeric con-stituents used in the fabrication of cell membranes (Eigen-brode, 2007). Not only do microorganisms inhabiting hy-drothermal systems have ready access to organics, they arealso supplied with abundant chemical energy provided bythe geochemical disequilibrium due to the mixing of hot hy-drothermal fluids and cold water. These energy-producingreactions are highly favorable for the kinds of microorgan-isms that obtain their energy from redox reactions involvinghydrogen or minerals containing sulfur or iron (Baross andDeming, 1995).

Another important aspect of the habitability of hy-drothermal systems is the ready availability of nutrients.High-temperature aqueous reactions leach volcanic rocksand release silica, Al, Ca, Fe, Cu, Mn, Zn, and many othertrace elements that are essential for microorganisms. Becausehydrothermal fluids are rich in dissolved minerals, they cre-ate conditions favorable for the preservation of biosigna-tures, i.e., traces of the life-forms that inhabit them. Althoughthe organic components of mineralized microfossils can beoxidized at higher temperatures (�100°C), more recalcitrantorganic materials (e.g., cell envelopes and sheaths) can betrapped and preserved in mineral matrices at lower tem-peratures [�35°C; Cady and Farmer (1996), Farmer (1999)]and thus allow chemical and isotopic analysis of organicbiosignatures. Minerals implicated in the fossilization of hy-drothermal microorganisms include silica, calcium carbon-ate, and iron oxide.

Some of the earliest life-forms on Earth might have in-habited hydrothermal environments (Farmer, 2000). Hyper-thermophiles occupy the lowest branches of the tree of life(Woese et al., 1990). Indeed, hydrothermal vent environ-

MEPAG ND-SAG500

ments, with their organic molecule-forming reactions, chem-ical disequilibria, and high nutrient concentrations, are con-sidered as a possible location for the origin of life (Russelland Hall, 1996). Some would argue, however, that the posi-tion of hyperthermophiles at the base of the tree of life is anartifact caused by the fact that such environments wouldhave represented protected habitats during the Late HeavyBombardment period, when a large part of the world’s oceanwas probably volatilized (Sleep et al., 1989). But the fact thathydrothermal environments could serve as protected habi-tats in hostile conditions is relevant to the early history ofMars.

Recently, it has been suggested that the suites of miner-als found at the surface of Mars (including silica and sul-fates) could be related to hydrothermal/fumarolic activity(e.g., Bishop et al., 2002; Squyres et al., 2007a, 2007b; Yen etal., 2007). Hydrothermal activity is to be expected becausevolcanic activity has occurred at the surface within the lastcouple of million years, which demonstrates that activeheat sources still exist (Neukum et al., 2004). In the eventthat life arose on Mars and flourished at the surface dur-ing the first 500 My of its history, the gradual deteriora-tion in surface conditions would have confined life-formsbeneath the surface, perhaps to be preserved in the cryos-phere and elsewhere. Conceivably, life might have adaptedto subsurface environments during the first 500 My andpersisted there since then. The subsurface environmentmight have sustained only very low rates of productivity,but it is also the most stable environment and a potentialhaven for life during large impacts. Volcanic activity in thevicinity of the cryosphere would lead to active hydrother-mal systems that, in some cases, might extend to the sur-face (Clifford, 1987).

The detection of hydrothermal activity on Mars is ex-tremely significant since these environments could representideal habitats for microorganisms that obtain their carbonand energy from inorganic sources. They might host extantlife as well as the fossilized traces of its ancestors. Returningintact samples of this lithology might be difficult for geo-logically recent material, which tends to be friable. It would,therefore, be very important to document the geologic con-text of such samples in case they do not survive the returntrip whole.

Criteria for sample size, selection, and acquisition proto-col would be the same as for the sedimentary suite. Exam-ples of possible lithologies for the hydrothermal suite includesamples from subsurface veins, fumarole deposits, surfacespring deposits from vent areas to distal apron environ-ments, as well as altered host rocks.

V-C. Low temperature altered rock suite

Low-temperature alteration processes occur at near-ambi-ent conditions on the martian surface (typically less thanabout 20°C) and include, among other things, aqueousweathering (including certain forms of palagonitization) anda variety of oxidation processes. Spectral observations madeby Viking and Pathfinder first inspired the notion that rocksurfaces on Mars are coated with thin veneers of altered ma-terial. Crude depth profiling provided by the RAT experi-ment on the MER rovers revealed thin (mm scale) alterationrinds on most rock surfaces studied. The exact nature of the

alteration processes remains under discussion, but most in-vestigators agree that low-temperature, relatively acidicaqueous conditions were involved (e.g., Haskins et al., 2005;Hurowitz et al., 2006; Ming et al., 2006).

Low-temperature processes also influence the regolithduring and after its deposition. The sulfur-rich compositionof regolith has long been attributed to low-temperature aque-ous processes that yielded sulfate and other secondary min-erals. This was confirmed when the MER rovers identifiedreactive magnesium and ferric sulfate minerals in the soils(Yen et al., 2007). The Viking gas exchange and labeled re-lease experiments also demonstrated that a reactive and ox-idizing compound in the regolith was capable of breakingdown many organic species. The nature and origin of thiscompound remains controversial, but various models call forlow-temperature processes, such as photochemical alter-ation, impact crushing, or oxidizing acid interactions (Yen etal., 2000; Hurowitz et al., 2007).

Understanding the conditions under which low-tempera-ture alteration processes proceed would provide importantinsight into the near-surface hydrological cycle, includingfluid/rock ratios, fluid compositions (chemical and isotopic,as well as redox conditions), and the mass fluxes of volatilecompounds (see also MacPherson et al., 2001, 2002). It wouldbe particularly important to analyze complete alteration pro-files, whether on rock surfaces or within regolith columns,because they would also constrain the kinetics of these al-teration reactions.

Representative, intact (or at least reconstructed) profileson rock surfaces would be required to understand these al-teration reactions. Recent experimental work has shown thatparent rock compositions (mineralogy) are an importantvariable in understanding these processes (Tosca et al., 2004;Golden et al., 2005). Consequently, a diverse compositionalsuite would be highly desirable and would require sample-site characterization during sample selection. Alteration pro-files on rock surfaces would most readily be acquired by cor-ing. The scales of alteration profiles range from less than 1mm to perhaps as much as 1 cm, so sample sizes of at least2 cm would be needed. Because alteration profiles are likelyto contain small amounts of sulfate and perhaps other reac-tive minerals, these samples would be susceptible to degra-dation during sampling and transport to Earth by processessuch as dehydration and chemical reaction, which in turncould also affect their physical integrity. Accordingly, sam-ple encapsulation is deemed critical.

V-D. Igneous rock suite

The igneous rocks on Mars are expected to be composedprimarily of lavas and shallow intrusive rocks of basalticcomposition (McSween et al., 2003; Christensen et al., 2005),along with volcanic ash deposits (e.g., Wilson and Head,2007). Although more- and less-evolved silicic and ultra-mafic magmatic rocks may potentially be present and wouldbe of great interest, they have not yet been unambiguouslyidentified on the surface. Igneous rocks would be central toinvestigations that reveal the geologic evolution of the mar-tian surface and interior because their geochemical and iso-topic compositions constrain both the composition of man-tle source regions and the processes that affected magmasduring their generation, ascent, and emplacement (see also

SCIENCE PRIORITIES FOR MARS SAMPLE RETURN 501

MacPherson et al., 2001, 2002). Although spacecraft instru-mentation could measure many major elements, Earth-basedanalyses of returned samples would be necessary to deter-mine most trace element and isotopic abundances of rocks.Melting and crystallization experiments in terrestrial labo-ratories would be based on the compositions of igneousrocks. Trace siderophilic element abundances and isotopiccompositions in igneous rocks could constrain the nature ofthe core and possibly its interaction with the mantle. Becausemagmas carried dissolved volatiles to the surface, these rockswould also be critical to understanding the inventories of de-gassed volatiles and the cycling of water and carbon.

Only igneous rocks could be reliably dated with absoluteradiometric dating techniques; therefore, they would be crit-ical for calibrating the martian stratigraphic timescale. Quan-tifying cratering rates would allow absolute ages of martiansurfaces to be derived from crater densities (Hartmann andNeukum, 2001). Unaltered igneous rocks that are geograph-ically linked to extensive terranes with known crater densi-ties would be required. This linkage would likely be ac-complished by comparing their geochemical/mineralogicalcharacteristics with local bedrock and by characterizing re-gional units with orbital remote sensing.

A large proportion of rocks on the martian surface arelikely to have experienced at least some low-temperature al-teration (Wyatt et al., 2004). Significantly weathered samples,however, would not satisfy the needs of these investigationsand, rather, would be better suited to investigations that in-volve rock/water interactions. Consequently, the low-tem-perature alteration products associated with the weatheringof the igneous rock suite are discussed separately.

To accommodate these investigations, a suite of igneoussamples with as much chemical and textural diversity aspossible would be required. Although some basaltic rocksmay appear similar in terms of major element abundancesand mineralogy, a suite collected over some geographicarea would be likely to exhibit differences in trace elementand isotopic compositions that would be highly informa-tive. If different types of igneous rocks are present, (e.g.,ultramafic or silicic rocks), additional samples of theserocks should be collected, as they could constrain frac-tionation processes on Mars. It is important to note thatmany different scientific objectives could be met with thesame samples. For example, radiometric dating of a lavaflow that overlies a sedimentary sequence might constrainthe cratering rate, the mechanisms and timing of planetarydifferentiation and evolution, and the period when sedi-mentation occurred. The igneous rock suite is relatively ro-bust; therefore, most geologic objectives could be met withminimal temperature control and encapsulation proce-dures. However, interactions with fluids derived from de-hydration of other samples, physical mixing, and the abra-sion of rock chips during transport could all be detrimentalto these investigations.

V-E. Regolith

The martian regolith reflects interactions between thecrust and the atmosphere, the nature of rock fragments,dust and sand particles that have been moved over the sur-face, H2O and CO2 migration between ice and the atmo-sphere, and processes that involve fluids and sublimation.Understanding regolith chemistry and mineralogy is vitalto determining the fates of any organic constituents. Someaspects of regolith studies necessarily overlap studies ofthe local rock petrology, geochemistry, and hydrothermaland low-temperature alteration processes. Althoughglobal-scale transport processes may have homogenizedmuch of the fine-grained martian regolith components, asshown by the similarity of most Viking and Pathfinder soilcompositions (e.g., Carr, 2006), the MER rovers have dem-onstrated that the regolith also contains a diverse range ofmineral assemblages, some of which originated locally.Other materials, such as volcanic ash (Wilson and Head,2007) and impact glass (Schultz and Mustard, 2004), mayhave come from greater distances. Understanding themechanisms by which these assemblages are produced isnecessary to understanding the evolution of the martiansurface and key fluid processes. The recent identificationof a silica-rich component in a Gusev Crater soil depositthat perhaps formed though hydrothermal processes (Ruffet al., 2007) and the presence of hematite spherules in theOpportunity soil (Squyres et al., 2004) highlight the im-portance of regolith studies. The mm scale alteration rindsidentified on rocks in the regolith in Gusev may have re-sulted from the reaction of S- and Cl-bearing species withminute amounts of liquid water (Haskins et al., 2005).Studying the mineralogy of alteration rinds within regolithgranules would give an insight into water and oxidationprocesses on Mars over long timescales (MacPherson et al.,2001).

A returned regolith sample would likely be evaluated inthe following way:

Size distribution studies of regolith particles may yield in-formation about local vs. distal provenance, as they didfor Apollo regolith samples (McKay et al., 1974).

Studies of regolith minerals and their morphology (with theuse of SEM, TEM, FTIR, and Raman spectroscopy tech-niques) and the chemistry of various lithologies within theregolith (scanning electron microscopy, transmission elec-tron microcopy, and EMPA) could help to quantify themobility of water, weathering processes, diagenesis, andchemical alteration in martian regolith, as has been donefor martian meteorites (Gooding et al., 1988; Velbel, 1988;Treiman et al., 1993) and Antarctic dry valley soils (Gib-son et al., 1983; Wentworth et al., 2005).

MEPAG ND-SAG502

FINDING: MSR would have its greatest value if the rocksamples are organized into suites of samples that rep-resent the diversity of the products of various planetaryprocesses. Similarities and differences between samplesin a suite can be as important as the absolute charac-

terization of a single sample. Four primary suites of rocksamples are called for:

• Sedimentary• Hydrothermal• Low-temperate water/rock products (weathering)• Igneous

Through studies of major elements and water soluble cations(Na�, K�, Ca2�) and anions (Cl�, SO42�, NO3�), the rela-tive extent and importance of the aeolian, salt-rich, sea-sonally active, and permanently frozen soil horizons couldbe determined and should be possible to evaluate for mar-tian regolith as well.

We already know that martian impact glasses containtrapped atmospheric gases (Bogard and Johnson, 1983),and the regolith could be an ideal sample in which to findthis component. Gas-release studies would be importantto interpret the history and evolution of the martian atmo-sphere.

Finally, a regolith sample would be used for toxicity tests,including intratracheal, corneal, dermal, and ingestionstudies.

The mixed and complex nature of regolith samples couldlead to unexpected findings. For example, Bandfield et al.(2003) proposed that atmospheric dust on Mars contains afew percent carbonate. This is important because carbonateprovides a record of atmosphere-water-crust interaction.However, carbonates have not yet been conclusively identi-fied on the surface of Mars, which makes the search for car-bonates within the dust from a regolith sample an importantcomponent for detailed mineralogical study. Microscopic ex-amination of the regolith sample in terrestrial laboratorieswould enable micrometeorites to be identified from whichmeteorite fluxes could be estimated.

A regolith sample is also likely to retain some CO2 andH2O. These might occur as ice or mixed clathrates. If ac-quired samples could be refrigerated at �10°C to �20°C, itmight be possible to identify their various potential species.Determination of CO2 and H2O abundance and isotopiccompositions would lead to a greater understanding of the global inventories and cycling between crust, atmo-sphere, and poles of these compounds. For example, accu-rate paleotemperatures of hydrothermal systems could be determined from measurements of 18O/16O isotopicfractionation during water-mineral isotopic exchange in hydrothermal assemblages (sampled across Mars or in me-teorites) by way of the isotopic analyses of martian ice asthe starting water-reservoir composition (Valley et al., 1997;Bridges et al., 2001). If a polar landing is not chosen, thenthe regolith sample would take on additional importanceas a likely source of the ice.

It is important to note that, for a geologic unit with a highpresumed degree of heterogeneity, like the martian re-golith, many of the measurements of interest could (andshould) be done in situ, and regolith studies should be animportant target for both landed missions and MSR. Thebasic field relationships, including measuring physicalproperties and their variation vertically and laterally,would best be done in place. However, sample returnwould be the best way to identify the altered and partiallyaltered materials, trace minerals (e.g., carbonates), rarelithologies, etc. It is also important to note that our experi-ence with the Spirit rover has shown us that we don’t havea good way of knowing the magnitude of geochemical/ge-ologic variability within this unit on a planetary scale andthe number of samples necessary to characterize it. This ob-jective should be thought of as one that would require morethan just the first MSR mission.

V-F. Polar ice

Samples of polar ice would be necessary to constrain thepresent and past climatic conditions on Mars as well as elu-cidate cycling of water. The samples necessary to achievethese objectives could include discreet samples of surface icefrom the PLD or a seasonal frost deposit. Short cores (�1 cmdiameter � 30 cm length) from the PLD or subsurface ice de-posit would also be desirable. A single sample could pro-vide critical input on surface/atmosphere interactions. Ashort core might resolve climatic variability in the last few100 Ka to 1 Ma (Milkovich and Head, 2005). Annual layerscould be observed in core samples, and isotopic signatures(�18O, D/H) are expected to define annual temperature vari-ability, changes in water reservoir availability and exchangewith the atmosphere, and short-term climate variations(Fisher, 2007). The composition of entrained non-ice dust ma-terials (e.g., aeolian, volcanic tephra, impact glass) wouldhelp determine the sources and relative proportions of dustthat reach the poles. Changes in the amount of entrainednon-ice dust with depth would help t