Treatises on the paleohydrology of Mars and Earth

Item Type text; Dissertation-Reproduction (electronic)

Authors Ferris, Justin Claus

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 06/10/2021 14:22:36

Link to Item http://hdl.handle.net/10150/289788 INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UIVll films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overiaps.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

ProQuest Information and Learning 300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA 800-521-0600

TREATISES ON THE PALEOHYDROLOGY OF MARS AND EARTH

by

Justin Claus Ferris

A Dissertation Submitted to the Faculty of the

DEPARTMENT OF HYDROLOGY AND WATER RESOURCES

In Partial Fulfillment of the Requirements for the Degree of

DOCTOR OF PHILOSOPHY WITH A MAJOR IN HYDROLOGY

In the Graduate College

THE UNIVERSITY OF ARIZONA

2002 UMI Number; 3050341

UMI ®

UMI Microform 3050341 Copyright 2002 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.

ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 2

THE UNIVERSITY OF ARIZONA ® GRADUATE COLLEGE

As members of the Final Examination Committee, we certify that we have read the dissertation prepared by JUSTIN CLAUS FERRIS entitled TREATISES ON THE PALEOHYDROLOGY OF MARS AND EARTH and recommend that it be accepted as fulfilling the dissertation requirement for the

Degree of DOCTOR OF PHILOSOPHY

Date 4/)102. Tom Maddock in Date

Ltherine K. Hirschboeck Date il >1 0^ Jon D. Pelletier Date V/ H/O 2. Robert G. Strom Date

Final approval and acceptance of this dissertation is contingent upon the candidate's submission of the final copy of the dissertation to the Graduate College.

I hereby certify that I have read this dissertation prepared xmder my direction and recommend that it be accepted as fiilfilling the dissertation requirement.

f/" /(JZ- Dissertation Director - Victor R. Baker Date 3

STATEMENT BY AUTHOR

This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations firom this dissertation are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained j&om the author.

SIGNED: Justin Claus Ferris 4

ACKNOWLEDGEMENTS

I sincerely thank James Dohm for introducing me to the geology of Mars. Without his mentoring, guidance, and support, this dissertation could not have been completed. I thank my advisor, Vic Baker, for allowing me the freedom to find my own path in graduate school, and for providing guidance, perspective, and a lifetime of philosophical insights. I thank the rest of my dissertation committee: Tom Maddock III, Katie Hirschboeck, Bob

Strom, and Jon Pelletier for always keeping their doors open to me. In particular, I want to thank Katie Hirschboeck for all the career advice and guidance she provided. I also thank

Tom Gehrels, Katie Hirschboeck, Bob Strom, Bob Clark, and Bob MacNish for providing me with opportunities to improve my teaching portfolio while at the same time collecting a badly-needed paycheck.

I would not have been able to achieve all that I have without the love and support of my parents, Ted and Claudia Ferris. They are all a son could ask for; words cannot express my gratitude for them. To my grandparents, Claus and Delores Holtrop, my sincerest gratitude for the financial and moral support you provided throughout my undergraduate and graduate education. To the rest of my family: Luke, Ted, and Beth Ferris, my thanks for being not only my siblings, but my best friends. Also, I wish to thank my new family, Larry,

Terri, T.T., and Lany Kraft, for your support and encouragement. Your enthusiasm for my accomplishments have made them that much more enjoyable.

Finally, I thank my best friend and wife, Gina Marie, for the endless well of love, support, strength, and understanding she has provided me in the almost six years we've been together. I could not have made it this far without you, nor could I make it a step further. 5

For Gina 6

TABLE OF CONTENTS

LIST OF ILLUSTRATIONS 10

ABSTRACT 11

CHAPTER 1.

INTRODUCTION 13

SYNOPSIS OF APPENDIX A 16

SYNOPSIS OF APPENDIX B 17

SYNOPSIS OF APPENDIX C 21

SYNOPSIS OF APPENDIX D 23

CHAPTER 2.

PRESENT STUDY 25

General Conclusions; Appendix A 25

General Conclusions: Appendix B 26

General Conclusions: Appendix C 26

General Conclusions: Appendix D 27

Summaiy of martian paleohydrology 28

REFERENCES 1 32

APPENDIX A.

ANCIENT DRAINAGE BASIN OF THE THARSIS REGION, MARS:

POTENTIAL SOURCE FOR OUTFLOW CHANNEL SYSTEMS AND

PUTATIVE OCEANS ORPALEOLAKES 35

PERMISSION 36 7

TABLE OF CONTENTS - Continued

ABSTRACT 37

INTRODUCTION 38

3-D PORTRAYAL OF THE EVOLUTION OF THE THARSIS

MAGMATIC COMPLEX 40

GEOLOGIC SUMMARY OF THE EVOLUTION OF THE

THARSIS MAGMATIC COMPLEX 43

BASIN MODEL: PHYSIOGRAPHIC SETTING AND

STRATIGRAPHIC SUMMARY 47

AQUIFER CHARACTERISTICS 50

DISCUSSION 53

ALTERNATIVE MODELS 56

CONCLUSION 59

REFERENCES 60

HGURES 70

APPENDIX B.

LATENT OUTFLOW ACTIVITY FOR WESTERN THARSIS, MARS:

SIGNIFICANT FLOOD RECORD EXPOSED 88

PERMISSION 89

ABSTRACT 90

INTRODUCTION 91

GEOLOGIC AND PHYSIOGRAPHIC SETTING 94 TABLE OF CONTENTS - Continued

STRATIGRAPHY

STRATIGRAPHIC SUMMARY OF THENSVs

DISCUSSION

PALEOHYDROLOGY

IMPLICATIONS

REFERENCES

FIGURES

APPENDIX C.

COMPARATIVE HYDROGEOMORPHOLOGY OF WET BEAVER

CREEK, ARIZONA, AND ABUS VALLIS, MARS

ABSTRACT

INTRODUCTION

WET BEAVER CREEK GEOLOGIC SETTING

HYDROGEOMORPHOLOGY AND PALEOHYDROLOGY OF

WET BEAVER CREEK

ABUS VALLIS GEOLOGIC SETTING

HYDROGEOMORPHOLOGY AND PALEOHYDROLOGY OF

ABUS VALLIS

DISCUSSION

CONCLUSIONS

ACKNOWLEDGEMENTS 9

TABLE OF CONTENTS - Continued

REFERENCES 169

FIGURES 173

APPENDIX D.

DARK SLOPE STREAKS ON MARS: ARE AQUEOUS PROCESSES

INVOLVED? 181

ABSTRACT 182

INTRODUCTION 183

GEOGRAPHIC AND GEOLOGIC SETTING OF THE DARK

SLOPE STREAKS 183

RATIONALE FOR THE RECONSIDERATION OF FLUVIAL

PROCESSES 184

SPRING DISCHARGE HYPOTHESIS 186

IMPLICATIONS 190

ACKNOWLEDGEMENTS 191

REFERENCES 192

HGURES 194

AUTHOR'S CURRICULUM VITAE 197 10

LIST OF ILLUSTRATIONS

Figure 1. Spatial relationships between features discussed in the manuscript 14

Figure 2. Temporal relationships between features discussed in the manuscript 15

Figure 3. Martian n-error Minimalization Criterion 20 11

ABSTRACT

A paleotopographic reconstruction has revealed the potential existence of an ancient enormous drainage basin and productive regional aquifer system in the eastern part of the Tharsis region. The basin model addresses a fundamental question in martian paleohydrology: Where did the massive amount of water required to carve the catastrophic outflow channels come from? These gigantic channels are typified in the recently discovered northwestern slope valleys (NSVs). The paleohydrologic implications for the NSVs are enormous, as they represent: (1) previously undocumented

Noachian and early Hesperian martian catastrophic floods, (2) a watershed to the northwest of Tharsis, possibly related to the early development of the circum-Chryse system of outflow channels, a watershed to the northeast that records flooding as recent as the early Amazonian, and (3) a potential source of water for a northern plains ocean and/or paleolakes. In addition to catastrophic outflows, Hesperian and Amazonian hydrological activity resulted in the formation of sapping valleys, such as Abus Vallis.

An investigation of Wet Beaver Creek, Arizona, revealed geological, hydrogeological, and geomorphic similarities between Wet Beaver Creek and Abus Vallis, Mars. As such, one may infer lithological and hydrological similarities as well. Wet Beaver Creek represents geological and climatological conditions that result in rapid, sometimes dynamic, fomiation of sapping channels. Sapping channels that occur along the highland-lowland boundary scarp, such as Abus Vallis, may now also be explained in terms of rapid fomiation during short-tenn periods of global climatic change, as well as by hypersaline aquifers and hydrothennal processes. The dark slope streaks on Mars may 12 represent yet another type of hydrogeomorphic process, different from the NSVs and

Abus Vallis. In light of Mars Global Surveyor-based geographic, geologic, and morphologic considerations, spring discharge is a viable explanation for the formation of dark slope streaks. Such activity would indicate the existence of near-surface aquifers and, more importantly, hydrological processes acting upon the surface of the planet, making it the only other hydrogeomorphically-active planet in the solar system besides

Earth. CHAPTER 1 - INTRODUCTION

This dissertation is non-traditional in the sense that it is composed of several published papers as opposed to a singular body of work. This option, offered by the

Graduate College, has afforded me an opportunity to complete my dissertation and, simultaneously, start building a publication record. A stipulation of this dissertation format is that I am required to explain how each of the following appendices reflects my unique contribution to the science of hydrology and, in relation to my co-authors, what portions of the manuscript represent my imique contribution to the body of research contained within.

I began working with Mr. James M. Dohm in the fall of 1999. Since then, I have worked closely and intensely with Mr. Dohm on a variety of projects, some of \Aiiich led to the manuscripts comprising this dissertation. I cannot honestly divide the intellectual work between his or my first-authored papers. How does one quantify iimumerable discussions, late-night brainstomaing, debates over yet another Starbucks cojBFee?

Therefore, I can only hope to modestly describe my contributions to the manuscripts, and hope I do not slight or offend any of the numerous people who made contributions to having these manuscripts successfully published. In the following paragraphs, I will summarize the contents of the manuscripts and describe "who did whaf for each of the appendices which compose this dissertation. The majority of the writing and assembly of the manuscripts contained in this dissertation were completed by Mr. Dohm and myself.

Furthermore, the hypotheses which compose the foundation of the two manuscripts first- 14 authored by Mr. Dohm were entirely his, and I was but fortunate to be able to contribute to such outstanding bodies of work.

Prior to any discussion of the manuscripts, it is important to establish the spatial and temporal relationships for the Tharsis Basin, the Northwestern Slope Valleys (NSVs),

Abus Vallis, and the Dark Slope Streaks (DSS). The above figure illustrates the approximate location of these features in relation to each other. The proposed boimdary 15 of the Tharsis Basin is illustrated in Figure 1 of Appendix A, while the global extent of the Dark Slope Streaks is illustrated in Figure 1 of Appendix D.

10000000000

1000000000 Northwestern Slope Valleys (NSVs) 100000000

10000000 •

1000000

100000

10000

BOBBOPOBOSODS 1000 Abus Valiis 100

asflaeaoOBooa 10 Dark Slope Streaksj

4.5 3.5 2.5 2 1.5 0.5 Billion Years Before Presont

The magnitude of the peak discharges and their occurrence in time is illustrated in the above figure. It is important to note that this is an approximation, as the duration of activity for the Northwestern Slope Valleys, Abus Vallis, and the Dark Slope Streaks is poorly constrained. Furthermore, the question marks above and below the map space occupied by Abus Vallis signify the large uncertainty in the peak discharges that occurred there. 16

SYNOPSIS OF APPENDIX A

This manuscript, entitled Ancient Drainage Basin of the Tharsis Region, Mars:

Potential Source for Outflow Channel Systems and Putative Oceans or Paleolakes, was first-avrthored by Mr. James M. Dohm and was co-authored by Dr. Victor R. Baker, Dr.

Robert C. Anderson, Mr. Trent M. Hare, Prof. Robert G. Strom, Dr. Nadine G. Barlow,

Dr. Ken L. Tanaka, Mr. Jim E. Klemaszewski, Dr. David H. Scott, and myself. The version herein was published in the Journal of Geophysical Research.

We created a paleotopographic reconstruction based on a synthesis of published geologic information and high-resolution topography that revealed the potential existence during &e Noachian Period of an enormous drainage basin and conciurent/subsequent aquifer system in the eastern part of the Tharsis region. Large topographic highs formed the margin of the gigantic drainage basin, which vras subsequently partly infilled by lavas, sediments, and volatiles. We hypothesize this produced an enormous and productive regional aquifer. The stacked sequences of water-bearing strata were then deformed locally, and, in places, exposed by magmatic-driven tq)lifts, tectonic deformation, and erosion. The basin model is important because it addresses a fundamental question in martian paleohydrology: Where did the massive amount of water required to carve the gigantic catastrophic outflow channels come from? Collectively, this model explains: (1) the paucity of Noachian outcrops and crustal magnetic anomalies in the proposed basin region, (2) the thick sequences of layered materials exposed in the walls of Valles

Marineris, (3) an anomalous concentration of layered ejecta craters in the Solis and

Thaumasia Planae regions, (4) the volatile-rich nature of Valles Marineris, and (5) the 17 observed hematite deposits detected in Valles Marineris. Most importantly, the basin model may provide a sufficient source of water necessary to carve the circum-Chryse outflow channel systems and the northwestern slope valleys (discussed in Appendix B), and to form putative northem-plains ocean(s) and/or paleolakes.

This manuscript resulted from the continuation of research conducted by Mr.

Dohm and Dr. Anderson (Dohm et al., 1998; Dohm and Tanaka, 1999; Anderson et al.,

2001). Mr. Dohm ibnnulated the initial hypothesis and was responsible for the bulk of the writing, creation of the graphics, and tibie assembly of the manuscript The various co­ authors contributed sections and/or graphics for the manuscript and/or provided critiques as to improve the manuscript My role, in general, was to work with the co-authors in bring^g out the hydrologic history of the Tharsis Magmatic Complex and Tharsis Basin hypothesis. Specifically, this included significant contributions to the "Aquifer

Characteristics" and "Alternative Models" sections and refining the hypotheses and text with Mr. Dohm.

SYNOPSIS OF APPENDIX B

This manuscript, entitled Latent outflofw activity for western Tharsis, Mars:

Significant flood record exposed, was first-authored by Mr. James M. Dohm and was co- authored by Dr. Robert C. Anderson, Dr. Victor R. Baker, Mr. Larry P. Rudd, Mr. Trent

M. Hare, Dr. Jim W. Rice, Jr., Mr. Robert R Casavant, Prof. Robert G. Strom, Dr. Jim R.

Zimbelman, Dr. David H. Scott, and myself. The version herein was published in the

Journal of Geophysical Research. 18

Observations permitted by the Mars Orbiter Laser Altimeter (MOLA) data revealed a system of gigantic valleys northwest of the huge martian shield volcano, Arsia

Mons, in the western hemisphere of Mars. We named these valleys the "northwestem slope valleys" (NSVs). ITie NSVs, which spatially correspond to gravity lows, are obscured by veneers of materials including volcanic lava flows, air fall deposits, and eolian materials. Geological investigations of the Tharsis region suggest that the system of gigantic valleys pre-dates the construction of Arsia Mons and its associated extensive lava flows of mainly Late Hesperian and Amazonian age. It also coincides stratigraphically with the early development of the Circum-Chryse outflow channels that debouch into Chryse Planitia. Although a multiple-working hypothesis for the formation of the NSVs was presented, structurally-controlled catastrophic flooding was deemed the most viable explanation. The paleohydrologic implications for the NSVs are enormous, as not only do they represent previously undocumented martian catastrophic floods, but they also imply a watershed to the northwest, where one had not been suspected before, perhaps related to the early development of the circum-Chryse system of outflow channels. Finally, they also provide another potential source of water for a northem plains ocean. The peak discharge rates have been estimated between about 10® and 10^° m /s. Even if maintained for only a short time, these discharges would have significantly contributed to the formation of a northem plains ocean.

This manuscript resulted from the continuation of research conducted by Mr.

Dohm and Dr. Anderson (Dohm et al., 1998; Dohm and Tanaka, 1999; Anderson et al.,

2001). Mr. Dohm formulated the initial hypothesis and was responsible for the bulk of 19

the writing, creation of the graphics, and the assembly of the manuscript. The various co­

authors contributed sections and/or graphics for the manuscript and/or provided critiques

as to improve the manuscript. My role, in general, was to develop a hydrologic history

for the northwestern slope valleys. Specifically, this included the contribution of the

"Paleohydrology" section. In order to generate this section, I had to devise a way to

calculate the potential discharges of this valley system. In the process, I created a unique

flow model that would calculate discharges based on assumed depths fix)m pre-formatted

cross-sections (as the raw data was unavailable at this time). This model was formulated

around a modified Manning's equation developed by Komar (1979).

I also developed a technique for selecting an appropriate Manning's n value for

the surface of Mars. I have dubbed it the "Martian n-error minimalization criterion".

Basically, it involves using the previously mentioned flow model (or any 1-D flow model) and calculating a variety of discharges for a variety of Manning's n values. An example is given below using the imlikely scenario where the entire northwestern slope valley

system was filled to the height of the mesas which contain them. Note that a range exists

where an equivalent difference in Manning's n makes far less difference in the calculated peak discharge. In this example, that range is pushed to the far right of the graph,

^jproximately 0.025 to 0.035. 20

Martian n-error MInlmallzation Criterion

4.00E+11

3.50E-<-11

3.00E+11

S 2.00E+11

1.50E+11

1.00E+11

5.00E+10 0.005 0.010 0.015 0.020 0.025 0.030 0.035 Value of Manning's n

However, for most of the other graphs (i.e., 50% filled, 10% filled) the "range of stability" lies closer to the middle, creating a "tipped-S" shape graph as the function begins to plimge downward again after leveling off. It is these areas of "leveling off", which I have referred to as the "range of stability", that lead me to select my Manning's n value. After averaging the graph values for fifty-plus fill calculations, a distinct range emerged where a relatively large (-0.005) deviation in Manning's n made little difference to the peak discharge. The center of this range was selected for use as my value of

Manning's n, 0.025. The ultimate goal was to select a reasonable value that, if it should deviate from the actual value, would make little difference in the calculation of the peak discharge. So it was surprising and rewarding to see that others had reached the same 21 conclusion as to the value of Manning's n for large martian discharges, using wholly different techniques than I had (Robinson and Tanaka, 1990).

Additionally, I created a table of comparative discharges (Table 1) and a graph of potential peak discharge for an assumed flow depth for the manuscript, as well as a scientific cartoon depicting late Noachian - early Hesperian activity (Figure 4). As with

Appendix A, I contributed to this manuscript in the refinement of the hypothesis and text with Mr. Dohm.

SYNOPSIS OF APPENDIX C

This manuscript, entitled Comparative Hydrogeomorphology of Wet

Beaver Creek, Arizona, and Abus Vallis, Mars, was first-authored by myself and co- authored by Mr. James M. Dohm and Mr. Trait M. Hare. The version herein has been submitted. At the time of this writing, the manuscript has not been officially accepted for publication.

Our detailed investigation of Wet Beaver Creek, Arizona, revealed an intriguing amount of geologic, hydrogeologic, and geomorphic similarities between Wet Beaver

Creek and Abus Vallis, Mars (and its siirrounding sapping channels). The channel floor of Wet Beaver Creek is dominated by well-imbricated cobbles and boulders, averaging between a half-foot and two feet in diameter. We determined these boulders were emplaced from multiple flood events, with at least one possessing a peak discharge of

42,000 cubic meters per second (1,500,000 cubic feet per second). Even more intriguing, we determined the mechanism for this extraordinary flooding to be predominately non- 22 darcian groundwater discharge. We attributed this to a unique set of geologic conditions, particularly that Wet Beaver Creek occurs along the Mogollon rim, a massive scarp and elevation dichotomy caused by the uplift of the Colorado Plateau, and that Wet Beaver

Creek occurs within a region of intersecting, deep-seated basement structure and fault swarms. Next, we then analyzed Abus Vallis using Mars Orbital Camera (MOC) imagery and Mars Orbital L^r Altimeter (MOLA) data This, coupled with previous investigations and geologic maps, allowed us to investigate the paleohydrology of Abus

Vallis and similar sapping channels in great detail. We determined Abus Vallis to be the result of similar hydrogeomorphic processes as Wet Beaver Creek. However, under current martian surface conditions, water is hydrodynamically unstable and Mars is theoretically covered in a planet-wide cryosphere. After examining multiple possible mechanisms for liquid water to be present in such an amount to allow for the formation of groimdwater sapping chaimels, we devised a multiple working hypothesis that included hypersaline aquifers, hydrothermal processes, and global climatic change. Given the similarity in genesis between Wet Beaver Creek and Abus Vallis, one may infer lithologic and hydrologic similarities as well. Wet Beaver Creek is an example of geologic conditions that allow the rapid, sometimes dynamic, creation of sapping channels. With this terrestrial analog, sapping channels that occur along the highland- lowland boimdary scarp, such as Abus Vallis, may now also be e?q)laiQed in terms of rapid formation during short-term periods of global climadc change, as well as by hypersaline aquifers and hydrothermal processes. Mr. Dohm introduced me to the field site, and the field work was conducted jointly by the two of us. Mr. Hare provided the Mars Orbital Laser Altimeter (MOLA) data that I needed for to create the HEC-RAS model of Abus Vallis and determine its slope with superior accuracy. Mr. Dohm assisted in refining the hypotheses and text I completed all other work related to the study and contained in this manuscript

SYNOPSIS OF APPENDIX D

This manuscript, entitled Dark Slope Streaks on Mars: Are Aqueous Processes

Involved?, was first-authored by myself and co-authored by Mr. James M. Dohm, Dr.

Victor R. Baker, and Dr. Tom Maddock HI. This manuscript has been acc^ted for publication in Geophysical Research Letters. The version herein is modified to contain material not published due to space limitations imposed by the journal.

Concentrations of dark slope streaks occur in the equatorial latitudes of Mars, mostly where magmatic-driven activity dominates the geologic record. Althou^ originally ascribed to wet debris flows, all the most recent published hypotheses concerning these features focus on processes which disturb a brighter dusty mantle to expose a darker substrate. These mechanisms invoke dry mass wasting or eolian processes, excluding a role for water. In light of the geographic, geologic, and morphologic considerations, and the new information provided from the Mars Orbital

Camera and the N^rs Orbital Data Altimeter, we reexamine fluvial processes as a viable explanation for some of the dark slope streaks. In our opinion, the two contrasting processes for the formation of dark slope streaks, dust avalanching and ^ring discharge. 24 may represent endpoints on a contmuum of progenitors. It may be that some of these features result from dry mass wasting or eolian processes, some from fluvial processes, and some from a mechanism(s) not yet conceived. Due to observations of dark slope streak features appearing on annual and decadal time scales, and their inferred fading over time, a spring discharge origin for the formation of the dark slope streaks has profound implications. This would infer that Mars has near-surface aquifers, but more importantly, that it has hydrologic processes acting upjon its surface, making it the only other hydrogeomoiphically-active planet in the solar system besides Earth.

I received helpful input from Mr. Dohm concerning maitian geology and refining the text Drs. Baker and Maddock provided insightful critiques to the manuscript I completed all other work related to the study and contained in this manuscript. 25

CHAPTER 2 - PRESENT STUDY

The methods, results, and conclusions of these studies are presented in the manuscripts appended to this dissertation. The following is a summary of the most important findings in these manuscripts, as well as a synopsis of the paleohydrology of

Mars in its entirety. A full treatise on the paleohydrology of Mars would take several lifetimes of study to complete. However, these manuscripts offer "windows" into the paleohydrology of Mars that, when combined with the research and teachings of others, particularly Dr. Victor R. Baker and Mr. James M. Dohm, knowledge of the relevant literature, and my own evolving thoughts and ideas, allow me to speculate on the "big picture" of martian hydrology.

Most scientists agree that very early in martian history. Mars was a planet with a hydrosphere much like the Earth's. Due to its small gravity well, its atmosphere was gradually lost to space, and as it cooled, the martian hydrologic cycle ceased. However, geomorphic evidence indicates that fluvial processes have been active at all scales upon the surface long after the martian hydrologic cycle had supposedly ceased. The existence of catastrophic outflow channels on northeastem Tharsis has long been recognized

(Baker, 1982), as well as the potraitial existence of paleolakes (Scott et al,, 1995) and oceans (Baker et al., 1991; Parker et al., 1993; Baker et al., 2000). What has been lacking is a coherent and self-consistent explanation for the extraordinary amounts of water required for all these features. Appendix A (Dohm et al., 2001a) provides this framework, describing the formation of an immense aquifer system where the Tharsis Magmatic

Complex now stands. This paper is eq)ecially critical to martian paleohydrology because 26 it provides for a large reservoir of water to be present in an area originally attributed to, and still believed by some, to be nothing more than a volcanic pile-up.

The circum-Chryse catastrophic outflow channels had long represented the largest flood events to occur in the solar system (Robinson and Tanaka). However, with the advent of MOLA data, a gigantic system of valleys was revealed to exist northwest of

Arsia Mons. Appendix B (Dohm et al., 2001b) details the discovery and interpretation of these valleys. The results, while not conclusive, favored the formation of these northwestern slope valleys by structurally-controlled catastrophic floods. The discovery of these valleys is significant in itself, as they may rq)resent the largest flooding ever to occur in the solar system. However, the implications of these valleys are just as important as their discovery. They may represent a northwest watershed for Tharsis, and therefore, imply that whatever mechanism led to the formation of all the catastrophic outflow channels of Tharsis was located centrally and perhaps to the south. Furthermore, this second set of valleys reduces the time and/or discharge necessary to fill ancient northern plains lakes and ocean(s), increasing the plausibility of the existence of both.

Not all channels on Mars are the result of catastrophic flooding. There exist numerous smaller sapping channels, formed at various times during the entirety of martian history. Clearly, some mechanism or mechanisms has allowed aqueous and fluvial processes to operate at conditions that are hydrodynamically unstable for water.

Appendix C (Ferris et al., submitted) attempts to clarify the potential mechanisms that could allow the observed geomorphology by drawing &om an Earth-based analog. This investigation of a carefiilly chosen valley system. Wet Beaver Creek, revealed geological. I

27

hydrological, and geomorphic characteristics surprisingly similar, yet xmique, when

compared to the sapping valley systems of the Colorado Plateau as described in the

literature. Yet, it closely simulates those same characteristics of martian sapping valleys

located in the northwest watershed of the Tharsis magmatic complex along the highland-

lowland dichotomy. The significance of this manuscript is two-fold; first, it shows that

groundwater-derived flooding may be more prevalent in terrestrial drainage systems than

originally thought Most research into groundwater-surface water relationships has

focused on losing streams or sapping chaimels. This observation underscores the utility

of paleoflood hydrology, including the identification and characteri2ation of

extraordinary flood events that many practicing hydrologists may overlook in the field.

Second, the Wet Beaver Creek field investigation of Ferris et al. (submitted) establishes

criteria for which a particular type of sapping channel, typified by Abus Vallis, may be

formed. In addition to processes previously discussed in the literature, such as gradual

formation over time due to hydrothermal systems (McKenzie and Nimmo, 1999; Ferris et

al., in press) and hypersaline aquifers (Knauth et al., 2000), this work raises the distinct

possibility that these channels could form during brief periods of climatic change (Baker

etaL, 1991,2000).

Appendix D (Ferris et al., in press) evaluates dark slope streaks and the possible

hypotheses that may explain their fomiation. Of the hypotheses that have been cited in

the literature, evidence collectively points toward die likely potential that modem-day

aqueous processes are acting upon the surface of Mars. As such, the dark slope streaks,

which have been observed to have fbmied since the arrival of the Mars Global Surveyor 28

(Edgett et al., 2000), may represent significant markers of hydrologically active regions on Mars. This significant finding severely conflicts with the paradigm that many planetary scientists are working within, that is, that Mars was once warm and wet but qmckly transitioned into a firozen desert, and has remained as such to this day.

To summarize, while humans had always envisioned Venus as our twin planet, it was in fact Mars that should have been given such a title. However, Mars is a twin as seen through a mirror, the reflection identical yet backwards as seen to the observer.

Where Earth is a hydrologically-active world punctuated by Ice Ages, Mars is a frozen world punctuated by periods of hydrologic activity (Baker et al., 1991, 2000; Baker

2001). In its earliest times, the martian atmosphere was thick enough to trap the heat of the fledgling Sun, and an Earth-like hydrologic cycle resulted. Mars' small size allowed for the lithosphere to thicken quickly, forever binding it into a monoplate world before flie end of the middle Noachian period (~3.9 Ga), and its low gravity allowed for its atmosphere to decay and an unknown amount of water to be lost to hydrodynamic escape. However, this series of events allowed for a new type of hydrologic cycle to arise, one not exogenically-driven by the Sun, but endogenically-driven by the planet's now-focused geothermal heat This heat forced its escape by moving magma throu^ deep-seated structural weaknesses, the predominant example being the infilling of the

Tharsis Basin and the creation of the Tharsis Magmatic Complex (Dohm et al., 2001a).

This volatile-rich region sat upon the planet's primary heat engine, a circumstance that would lead to the creation of the largest flood features in the solar system, the circum- 29

Chryse outflow channels (Baker, 1982; Robinson and Tanaka, 1990) and the northwestern slope valleys (Dohm et al., 2001b).

Planet-wide formation of sapping channel formation throughout geologic time has resulted from a combination of processes. Undoubtedly, hydrothermal activity has been, and is currently, occurring on the surface of Mars (i.e., Newsom, 1980; McKenzie and

Nimmo, 1999). Another process for advancing sapping channel growth is simply to have climatic conditions that allow liquid water to be stable at the martian surface. The previously-mentioned processes of punctuated magmatic activity causing catastrophic flooding also results in an additional effect on Mars: There is enough volcanic greenhouse gas release to dramatically warm the martian climate (Baker et al., 2000;

Dohm et al., 2000). The brief periods of climatic change (Baker, 2001), analogous in some ways to Earth's Ice Ages, allow a strengthened hydrological cycle to exist, moving more water through the atmosphere, creating and/or enlarging glaciers (Kargel et al.,

1995), and, perhaps more infrequently, creating a vast ocean and/or lakes in the northern hemisphere (Baker et al., 1991, 2000). These periods of climatic change allow for sapping channel growth, and, as illustrated by Ferris et al. {submitted)-, this growth may be rapid and allow for the extensive advancement of sapping channels during brief periods of climatic change. Finally, another mechanism exists which has undoubtedly played a role in sapping channel growth, and tiiat is the presence of brines and hypersaline waters at the martian surface. The Earth has had a vigorous hydrologic cycle for approximately 4.5 billion years, yet in that time, it has produced only 2.77% fresh water (of Earth's total water budget), and 77% of that is locked up in ice caps and 30 glaciers (Fetter, 1994). Considering Mars' Earth-like hydrologic cycle ceased early in its history, and it has experienced only geologically-brief bouts of climatic change since then, its hard to imagine a mechanism which would allow large quantities of fresh water to exist. Furthermore, in arid regions, a distinction that Mars certainly qualifies for, the slow circulation of groundwater can result in extensive mineralization. Given these two factors, it seems likely that Mars is a planet of groundwater brines, Knauth et al. (2001) noted that water could exist in liquid form at the martian surface under current temperature and pressures if the water was sufficiently rich in dissolved salts. This salinity, regardless of its severity, would aid both hydrothermal systems and global climatic change in the creation of liquid water at or near the martian surface.

Aside from catastrophic outflow channels and groundwater sapping channels.

Mars also has large drainage systems and evidence of spring discharge (Ferris et al., in press). The large drainage networks (i.e., Warrego Valles) either result from ancient spring discharge related to hydrothermal activity (Gulick, 1993; Dohm et al., 2001c) or perh^s precipitation that occurred during a period of martian climatic change. However, there exist more recent features (i.e., dark slope streaks) that may result from modem-day aqueous processes acting iqxjn the martian surface. Clearly, the complexity of martian paleohydrology is greater than previous researchers had considered (Clifford, 1993). The

Earth is best described as a "water world", with approximately 70% of its surface covered in water. However, approximately 30% of its land surface can be classified as a desert

As more data is collected, and more research is done, I believe Mars will reveal an 31 e(]ually complex story of the allocation, movement, and deposition of its water over geological time. 32

REFERENCES

Anderson, R. C., J. M. Dohm, M. P. Golombek, A. Haldemann, B. J. Franklin, K. L. Tanaka, J. Lias, and B. Peer, Primaiy centers and secondary concentrations of tectonic activity through time for the western hemisphere of Mars, J. Geophys. Res., 106,20,563-20,585,2001.

Baker, V. R., The channels of Mars, Aijstin, University of Texas Press, 198 pp, 1982.

Baker, V. R., Water and the Martian landscape. Nature, 412,228-236,2001.

Baker, V. R., R. G. Strom, V. C. Gulick, J. S. Kargel, G. Komatsu, and V. S. Kale, Ancient oceans, ice sheets and the hydrological cycle on Mars, Nature, 352,589-594, 1991.

Baker, V. R., R. G. Strom, J. M. Dohm, V, C. Gulick, J. S. Kargel, G. Komatsu, G. G. Ori, and J. W. Jr. Rice, Mars' Oceanus Borealis, ancient glaciers, and the MEGAOUTFLO hypothesis. Lunar Planet Sci. Conf. [CD-ROM], XXXI, abstract 1863,2000.

Clifford, S. M., A model for the hydrologic and climate behavior of water on Mars, J. Geophys. Res., 98,10,973-11,016,1993.

Dohm, J. M., R. C. Anderson, and K. L. Tanaka, Digital structural mapping of Mars, Astronomy & Geophysics, 39,3.20-3.22,1998.

Dohm, J. M., and K. L. Tanaka, Geology of the Thaumasia region. Mars: plateau development, valley origins, and magmatic evolution. Planetary and Space Science, 47,411-431,1999.

Dohm, J. M., R. C. Anderson, V. R. Baker, R. G. Strom, G. Komatsu, and T. M. Hare, Pulses of magmatic activity through time: potential triggers for climatic variations on Mars, Limar Planet. Sci. Conf. [CD-ROM], XXXI, abstract 1632,2000a. 33

Dohm, J. M., J. C. Ferris, V. R. Baker, R. C. Anderson, T. M. Hare, R. G- Strom, N. G. Barlow, K. L. Tanaka, J. E. Klemaszewski, and D. H. Scott, Ancient Drainage Basin of the Tharsis region. Mars: Potential source for outflow channel systems and putative oceans or paleolakes, Journal of Geophysical Research, Vol. 106, p. 32,943- 32,958,2001a.

Dohm, J, M., R. C. Anderson, V. R. Baker, J. C. Ferris, L. P. Rudd, T. M. Hare, J. W. Rice, Jr., R. R. Casavant, R. G. Strom, J. R. Zimbelman, and D. H. Scott, Latent outflow activity for westem Tharsis, Mars: Significant flood record exposed. Journal of Geophysical Research, Vol. 106,12,301 -12,314,2001b.

Dohm, J.M., K. L. Tanaka, and T. M. Hare, Geologic map of the Thaimiasia region of Mars: USGS Misc. Inv. Ser. Map 1-2650, scale 1:5,000,000,2001c.

Edgett, K. S., M. C. Malin, R. J. Sullivan, P. Thomas, and J. Veverka, Dynamic Mars: new dark slope streaks observed on annual and decadal time scales. Lunar Planet. Sci. Conf. [CD-ROM], XXXI, abstract 1058,2000.

Ferris, J.C., J. M. Dohm, V. R. Baker, and T. Maddock III, Dark Slope Streaks on Mars: Are Aqueous Processes Involved?, accepted for publication in Geophysical Research Letters.

Ferris, J. C., J. M. Dohm, and T. M. Hare, Comparative Hydrogeomorphology of Wet Beaver Creek, Arizona, and Abus Vallis, Mars, submitted.

Fetter, C. W., Applied Hydrogeology, Englewood Cliffs, Prentice-Hall, 1994.

Gulick V. C., Magmatic intrusions and hydrothermal systems: Implications for the formation of small Martian valleys, Ph.D. Thesis, Univ. of Arizona, Tucson, Arizona, 1993.

Kargel, J. S., V. R. Baker, J. E. Beget, J. F. Lockwood, T. L. Pewe, J. S. Shaw, and R. G. Strom, Evidence of ancient continental glaciation in the martian northern plains, J. Geophys. Res., 100,5,351-5,368,1995. 34

Knauth, L., P., D. M. Bmt, and S. Klonowski, Dense, eutectic, valley-forming, intermediate latitude (DEVIL) brines on Mars, Geol. Soc. America Abstracts with Pro^ams, v. 32, no. 7, p. A303,2000.

Komar, P. D., Comparisons of the hydraulics of water flows in Martian outflow channels with flows of similar scale on Earth, Icarus, 37,156-181,1979.

McKenzie, D., and F. Nimmo, The generation of Martian floods by the melting of ground ice above dykes. Nature, 397,231-233, 1999.

Newsom, H. E., Hydrothermal alteration of impact melt sheets with implications for Mars: Icarus, 44,207-216,1980.

Parker, T. J., D. S. Gorsline, R. S. Saunders, D, C. Pieri, and D. M. Schneeberger, Coastal geomorphology of the martian northern plains, J. Geophys. Res., 98, 11,061-11,078, 1993.

Robinson, M. S., and K. L. Tanaka, Magnitude of a catastrophic flood event at Kasei Valles, Mars, Geology, 18,902-905,1990.

Scott, D. H., J. M. Dohm, and J. W. Jr. Rice, Map of Mars showing channels and possible paleolake basins, USGSMisc. Inv. Ser. Map 1-2461 (1:30,000,000), 1995. 35

APPENDIX A

ANCIENT DRAINAGE BASIN OF THE THARSIS REGION, MARS: POTENTIAL SOURCE FOR OUTFLOW CHANNEL SYSTEMS AND PUTATIVE OCEANS OR PALEOLAKES.

J.M. Dohm, J.C. Ferris, V.R. Baker, R.C. Anderson, T.M. Hare, R.G. Strom, N.G. Barlow, K.L. Tanaka, J.E. Klemaszewski, D.H. Scott

Manuscript published in the Journal of Geophysical Research, Volume 106, Pages 32,943-32,958, December 2^, 2001. 36

FIRST AUTHOR PERMISSION

I hereby grant permission to Mr. Justin C. Ferris to include the following manuscript in this dissertation submitted to the faculty of the Department of Hydrology and Water Resources and the Graduate College of the University of Ari2»na.

Manuscript Title; Ancient Drainage Basin of the Tharsis Region, Mars: Potential Source for Outflow Channel Systems and Putative Oceans or Paleolakes.

James M. Dohm 37

ANCIENT DRAINAGE BASIN OF THE THARSIS REGION, MARS:

POTENTIAL SOURCE FOR OUTFLOW CHANNEL SYSTEMS AND

PUTATIVE OCEANS OR PALEOLAKES.

J.M. Dohm\ J.C. Ferris^ V.R. Baker''^, R.C. Anderson^, T.M. Hare'*, R.G. Strom\ N.G.

Barlow^, K.L. Tanaka'*, J.E. Klemaszewski^ D.H. Scott^ 'Department of Hydrology and

Water Resources, University of Arizona, Tucson, AZ, 85721, imd@,hwr.arizona.edu,

^Lunar and Planetary Laboratory, ^Jet Propulsion Laboratory, Pasadena, CA, University of Arizona, Tucson, AZ, \f.S. Geological Survey, Flagstaff, AZ, University of Central

Florida, Orlando, FL, ^Arizona State University, Tempe, AZ, ^In Memory of a Geologist,

ABSTRACT

Paleotopographic reconstructions based on a synthesis of published geologic information and high-resolution topography, including topographic profiles, reveal the potential existence of an enormous drainage basin/aquifer system in the eastern part of the Tharsis region during the Noachian Period. Large topographic highs formed the margin of the gigantic drainage basin. Subsequently, lavas, sediments, and volatiles partly infilled the basin, resultmg in an enormous and productive regional aquifer. The stacked sequences of water-bearing strata were then deformed locally, and, in places, exposed by magmatic-driven uplifts, tectonic deformation, and erosion. This basin model provides a potential source of water necessary to carve the large outflow channel systems of the Tharsis and surrounding regions, and to contribute to the formation of putative northern-plains ocean(s) and/or paleolakes. 38

INTRODUCTION

Stratigraphic, tectonic, and erosional records, compiled through geological mapping investigations at regional and local scales, demonstrate a significant contribution of magmatic-driven processes to the dynamic geologic history of Mars [for example, Mouginis-Mark, 1985; Scott and Tanaka, 1986; Greeley and Guest, 1987;

Baker et al, 1991; Crown et al., 1992; Scott et al, 1993; Robinson et al, 1993; Crown and Greeley, 1993]. These processes are perhaps best exemplified at Tharsis and the surrounding regions in the western hemisphere \Frey, 1979; Wise et al, 1979; Plescia and Saunders, 1982; Solomon and Head, 1982; Scott and Tanaka, 1986; Morris and

Tanaka, 1994; Scott and Zimbelman, 1995; Anderson et al, 1998, JGR Planets - accepted; Scott et al, 1998; Golombek, 2000; Solomon and Head, 2000], where pulses of magmatic activity associated with the development of the Tharsis magmatic complex

(TMC) [e.g., Dohm et al, 2000a; Dohm et al, 2000b] may have triggered catastrophic floods and short-lived climatological perturbations [Baker et al, 1991; Baker et al,

2000; Baker, 2001].

The Tharsis magmatic complex is comprised of numerous components that formed during specific stages of the complex's development (Figures 1 and 2), including volcanic constructs of varying sizes and extensive lava flow fields, large igneous plateaus, feult and rift systems of varying extent and relative age of formation, gigantic outflow channel systems, vast canyon systems, and local and regional centers of tectonic activity. Many of the local and regional centers of tectonic activity [Anderson et al,

1998; Anderson and Dohm, 2000; Anderson et al, JGR Planets - in press] are 39

interpreted to be the result of magmatic-related activity, including uplift, faulting, dike

emplacement, volcanism, and local hydrothermal activity [Dohm et al, 1998; Dohm and

Tanaka, 1999; Dohm et al, 2000b,d\. The geologic history of the numerous components

of the Tharsis magmatic complex, which has been analyzed in detail by numerous

investigators (see Figure 1), is incorporated into the analyses and interpretations that

follow.

In as much as the relative age of formation of flie primary components of the

complex are generally understood, the general stratigraphy established, and the high-

resolution topography of present-day Mars known, the features and materials can be

"backstripped" systematically from their present configuration. This provides an

approximate sequential 3-D view of the paleotopography at the cessation of each of the

five stages of the complex's development (Figures 3-7). Stage information correlates

with the scheme devised by Dohm and Tanaka [1999] (see Figures 1 and 2).

The purpose of the 3-D visualization and quasi-quantitative backstripping is not to

generate a quantitatively accurate reconstmction of martian paleotopography at discrete

time steps. Such a reconstmction may well be possible at a fiiture date when more data

become avail^le. However, the more qualitative visualizations employed in this study

serve to summarize inferences made fix}m geological mapping and analyses and synthesis

of published map information. This summary, really an illustrated working hypothesis, leads to the identification of an ancient, gigantic draine^e basin that persists throi^

much of the history of the region. Moreover, the drainage basin, once identified, is foimd

to be consistent with a diverse series of other observations. The overall coherence of 40 these observations can be considered to be a tentative conjSrmation of utility for the basin hypothesis. While the quantitative confirmation of detailed validity for various reconstructions is beyond the scope of this preliminary study, such reconstructions would be among the several possible follow-ons to the present study.

3-D PORTRAYAL OF THE EVOLUTION OF THE THARSIS MAGMATIC

COMPLEX

A 3-D visualization program (Bryce 4 by the MetaCreations Corporation) provides a heuristic representation of the geological evolution of the Tharsis magmatic complex. Although this program cannot modify present-day MOLA-based topography in absolute elevation values, it can mold surfaces and render and animate 3-D scenes.

Visual appearances for each of the five major stages of activity of the Tharsis magmatic complex were generated using the above-referenced topographic, stratigraphic, paleoerosional, and paleotectonic information. This information includes spatial and temporal relations among shield volcanoes, igneous plateaus, magmatic-driven centers of tectonic activity, fault and rift systems, and lava flow fields. In this process, features and materials of a known relative age were systematically "backstripped" fiwm their present

MOLA-based configuration, providing approximate sequential views of the paleotopogr^hy at the cessation of each of the five stages of the complex's development

We emphasize that these illustrations are qualitative at this point They represent a working hypothesis, based on our interpretation of the geologic mapping and analyses and synthesis of published m^ information, that should be tested with quantitative data 41 as these become more available in formats that can be related to the geology. We also emphasize that the thicknesses and aerial extent of features/materials are approximated, because geologic histories for Mars are established through photogeologic mapping and relative-age determination of surfaces and structures. Even on Earth, paleotopographic reconstructions are difiGcult where a much richer set of tools for analyzing the geologic record can be used, including field mapping and absolute dating of rocks.

In addition to utilizing published geologic information and cross sections (e.g..

Figure 8a,b,c), our reconstructions make use of MOLA-derived topographic projSles (e.g..

Figures. 9-11) to best estimate thicknesses and aerial extents of features and materials.

Figures 9 and 10, for example, show substantial rises in the central and western parts of

Valles Marineris, which are located near the central part of the proposed drainage basin.

These rises, which have been identified as centers of tectonic activity [Anderson et al,

1998, JGR Planets - accepted; Anderson and Dohm, 2000], are interpreted to be the resxilt of magmatic-driven uplifts and local volcanism and hydrothermal activity along large structural discontinuities in the martian crust [Dohm et al., 1998; Anderson and

Grimm, 1998]; the Late Noachian-Hesperian centers of tectonic activity post date the proposed drainage basin.

We started from Stage 5 (present day—Amazonian Period; Figure 3) by importing the MOLA Experiment Gridded Data Record (EGDR) and ended with Stage 1 (Noachian

Period; Figure 7). For example, the Late Hesperian/Amazonian (Stages 4-5; Figures 3 and 4, respectively), gigantic volcanic constructs of Tharsis Montes and Olympus Mons and their associated lavas flow fields were removed from the Noachian/Early Hesperian 42

(Stages 1-3) reconstructions using Bryce surface editing tools (Figures 5-7, respectively).

In addition, materials of regional extent with no obvious point source (e.g., volcanic construct) were removed fi-om the topographic reconstructions of earlier stages of the complex's development using published geologic map information coupled with MOLA- based geologic cross sections. For example. Early Hesperian (Stage 3) ridged material was removed from the Noachian reconstmction (Stage 1 and early part of Stage 2) using the geologic map of the western hemisphere [Scott and Tanaka, 1986\ and MOLA-based geologic cross sections (e.g.. Figure 8a,b,c). In other situations, materials were added to approximate the original topography that may have been subsequently wom down by erosional processes for each stage of the TMC's development For example, because

Claritas rise developed diiring the Noachian Period [Anderson et al, 1998; JGR Planets - accepted] and was subsequently modified by erosional and tectonic process, material was added to the rise to approximate its original configuration during Stage 1 (Figure 7).

Again, we wish to reemphasize that this methodology is qualitative, based on experience with terrestrial and martian geology. Although the details (e.g., unit thicknesses) will be refined as our knowledge of Mars increases, the major features and sequences of events presented here provide an improved understanding of the evolution the Tharsis magmatic complex, as well as its influence on global geology and paleoclimate. 43

GEOLOGIC SUMMARY OF THE EVOLUTION OF THE THARSIS

MAGMATIC COMPLEX

This simimaiy provides an interpretive geologic history of the Tharsis Magmatic

Complex based on topographic, stratigraphic, paleoerosional, and paleotectonic information compiled from the work of numerous investigators. Key information used in this construction includes the spatial and temporal relations among shield volcanoes, igneous plateaus, magmatic-driven centers of tectonic activity, fault and rift systems, and lava flow fields (see Figures 1 and 2). Stage information is based on Dohm and Tanaka

[7PPP] and roughly corresponds to the martian stratigraphic scheme [Tanaka, 1986].

Early to Middle Noachian, Part of Stage 1 (Figures J, 2, and 7). The greatest percentage of faults preserved in Noachian materials of the western hemisphere originate near the central part of the Claritas rise. This region is marked by an enormous rifl system and hi^y-deformed promontories interpreted to be basement complex. The

Claritas rise is a center of activity representing a region of broad magmatic-driven uplift and associated volcanism and tectonism. In addition to Claritas, magmatic-driven tectonic activity is also identified for the Tempe plateau and pre-Tharsis Montes rises:

Uranius, Ceraunius, and Arsia SW. Uncertainty exists in the commencement of ancient local and regional centers of magmatic-related activity. Syria Planum and Arsia rise, for example, most likely began as local centers of activity during the Early to Middle

Noachian with substantial growth, perhaps episodically, through the Hesperian. The

Noachian magmatic activity mostly occurs along large fiacture/fault zones, many of which may represent large dislocations in the martian crust/lithosphere. Such 44 dislocations may be the result of the TMC development and (or) represent plate or block boundaries formed dnring the period of high heat flow [e.g., Schubert et al, 1992; Sleep,

1994].

Broad rises, rugged mountain ranges, and a ridge of materials, which may represent the remains of a highly eroded rim of Chryse impact basin, partly form the margin of the proposed enormous Noachian drainage basin. The development of the highland-lowland boundary sometime during the Middle to Late Noachian may have resulted in a substantially different paleohydrologic regime in Chiyse Planitia region, including an enhanced hydraulic gradient

Late Noachian to Early Hesperian, Stage 2 (Figures i, 2, and 6). In addition to continued growth of the Arsia rise (pre-Tharsis Montes), centers of magmatic-driven tectonic activity and possible associated volcanic eruptions and hydrothermal activity are identified near the central part of Valles Marineris, Syria Planum, and the source region of Warrego Valles. Numerous faults, for example, are radial to or concentric about the central part of Valles Marineris, representing a broad center of magmatic-driven uplift along a large crustal/lithospheric dislocation (Figure 9). Similar to the central part of

Valles Marineris, Syria Planum is also a site of long-lived (Noachian through at least

Late Hesperian) magmatic/tectonic activity, ^^ch includes domal uplift and associated radial and concentric faulting, but at a much larger scale and longer duration than is recognized at the central part of Valles Marineris.

Magmatic-related activity such as doming xmderlying Arsia Mons and at Syria

Planum and central Valles Marineris may be genetically associated with the early development of the circxim-Chryse outflow channel systems [e.g., Dohm et al., 1998;

McKemie and Nimmo, 1999] as well as the formation of the newly defined northwest slope valleys [Dohm et al., 2000c, 2001], located on the opposite side of the Tharsis rise from the circimi-Chryse system of oxitflow channels, several thousands of kilometers to the west In addition, the source region of Warrego Valles has been interpreted to be a site of intrusive-related doming and tectonic and hydrothermal activity resulting in the formation of well-defined valley networks of Warrego Valles [Gulick, 1993; Dohm and

Tanaka, 1999].

The Thaumasia plateau uplift also occurred during this time [Dohm and Tanaka,

1999]. The plateau uplift and local/regional centers of magmatic activity largely modified the paleotopography of the TMC regioa These activities resulted in the masking of the Tharsis basin, releasing catastrophic floods that led to the early formation of the Circum-Chryse system of outflow chaxmels [e.g., Rotto and Tanaka, 1995; Nelson and Greeley, 1999] and the northwest slope valleys. This probably drove volatiles (e.g., groimdwater) fi-om uplifted regions such as the central part of Valles Marineris and Syria

Planum into nearby topographic lows [e.g., Dohm et al, 2000b; Barlow et al, 2001].

The time of initial formation of these outflow channel systems is uncertain (Figure 2).

For example, lava flows sourcing fi*om Stage 1 and Stage 2 centers of activity may have partly infilled and followed paleovalleys associated with early development of the large outflow channel systems resulting in an inversion of topography (lava ridges/mesas resulting fix>m subsequent erosion of less competent brecciated surrounding materials).

Later episodes of magmatic-triggered flooding may have formed new valleys or may 46 have followed paleovalleys. Also during this time, extensive older ridged plains materials [Dohm and Tanaka, 1999] and intercrater materials were emplaced in topographically low areas. Some intercrater materials may be the result of phreatomagmatic explosions, such as in fte Valles Maiineris region where magma/water/water-ice interactions have been proposed [Chapman et al, 2001].

Early Hesperian^ Stage 3 (Figures 1, 2, and 5). Continued magmatic/tectonic activity is identified at Arsia-SW dome (continued growth of Tharsis rise drainage divide), Syria Planimi, and Tempe plateau. Incipient activity (pre-Tharsis Montes activity) is also recognized in the Pavonis Mons [Plescia and Saunders, 1982; Anderson et al, 1998, JGR Planets - accepted] and Alba and Ulysses Patera regions. Also during the Early Hesperian Period, additional development is recorded for the circum-Chryse system of outflow channels and at the northwestern slope valleys, possibly related to another pulse of magmatic activity at Arsia-SW dome, Syria Planum, and the central part of Valles Marineris. Extensive younger ridged plains materials were emplaced in topographic lows.

Late Hesperian to Early Amazonian, Stage 4 (Figures ly 2, and 4). Significant volcanic activity is also recorded during the Late Hesperian and Early Amazonian, including the development of Olympus Mons and the Tharsis Montes shield volcanoes.

Also emplaced were voluminous sheet lavas centered at the large shield volcanoes and at

Syria Planum. Lava flows centered at Arsia Mons, for example, may extend to the northwest as far as the northwestern slope valleys embaying the gigantic northwest- trending promontories and partly infilling the system of valleys [Zimbelman et al., 2000]. 47

A significant transition from magmatic-tectonic activity to volcanic activity is observed at the major centers of activity, notably at Syria Planum [Dohm and Tanaka,

1999]. The final appearance of large-scale tectonism for the western hemisphere is associated with the dominant center of tectonic activity. Alba Patera [Anderson et al,

1998]. Significant outflow channel formation also occurred in the circum-Chryse system of outflow channels during the Late Hesperian and into the Early AmazoniaiL This may be related to yet another pulse of magmatic activity at Tharsis Montes, Syria Planum, and the central part of Valles Marineris, which is consistent with the driving mechanism in the MEGAOUTFLO hypothesis [Baker et al, 1991, 2000].

Amazonian, Stage 5 (Figures 1, 2, and 3). Evidence for continued growth of

Olympus Mons and The Tharcis Montes shield volcanoes is identified during this geologic period. Isolated occurrences of magmatic/tectonic activity appear to be associated with continued construction of the Tharsis Montes shield volcanoes and

Olympus Mons. Other than minor graben formation associated with the final stages of

Alba Patera, these local volcanic sources may represent late-stage pulses of magmatic- tectonic activity in the Tharsis region.

BASIN MODEL: PHYSIOGRAPHIC SETTING AND STRATIGRAPHIC

SUMMARY

During the Early to Middle Noachian (early part of Stage 1), topographic highs associated with centers of tectonic activity (Uranius Fossae, Ceranius Fossae, Arsia and

Claritas rises), rugged mountain ranges of the east-trending Thaumasia highlands and 48 north-trending Coprates rise, high-standing terrain located southwest of Chryse Planitia

(interpreted here to be a highly-eroded impact crater rim), and Tempe plateau formed the margins of the proposed drainage basin (Pigure 7). During this same time period, lavas, sediments, and volatiles partly infilled the basin resulting in a large aquifer system. The partial infilling of the basin perhaps coincides with reported higher rates of planet-wide surface degradation than for post-Noachian time \Masurs1

Barlow, 1990; Craddock and Maxwell, 1993; Scott et al, 1995; Carr and Chaung, 1997;

TanakaetaL, 1998\.

The basin/aquifer system was subsequently obscured and deformed by middle

Noachian and younger geologic activity (Stages 1-5), including:

(1) the Thaumasia plateau uplift,

(2) continued growth of the Arsia rise and the development of other centers of

tectonic activity, which also correspond to topographic rises (including central

Valles Marineris rise, Syria Planum, and pre-Tharsis-Montes Pavonis),

(3) the emplacement of the intercrater plains materials, older and younger ridged

plains materials, and lavas associated with Syria Planum and Tharsis Montes

and local sources such as fissure fed lavas, and

(4) the construction of the Tharsis Montes shield volcanoes.

The interaction of magmatic intrusions with water-bearing strata may have resulted in lateral migration of subsurface volatiles away from magmatic-driven heating and uplift

Such occurrences were probably common throughout the evolution of the drainage basin

[Anderson et al, 1998, JGR Planets - accepted; Dohm et al., 1998; Dohm et al, 2000e; 49

Solomon and Head, 2000] and may explain the concentrations of near-surface volatiles in the Solis Planum region expressed by an anomalous concentration of layered ejecta craters [Barlow et al, 2001]. In addition, stacked sequences of basin materials were locally exposed in the canyon walls of Valles Marineris by magmatic-driven doming, tectonic deformation, and erosion.

Whether the proposed drainage basin is strictly the result of surrounding topographic highs or whether there is an endogenic component is difScult to answer. As noted in the previous section, the central part of the proposed basin, Valles Marineris, is the location of centers of tectonic activity. These centers are interpreted to be the result of magmatic-driven uplifts and local volcanism, dike emplacement, and hydrothermal activity. In addition, more than half of the proposed basin was modified by the

Thaumasia plateau uplift, which may have been the result of magma plume head. Such activity may be likened to topographic inversions observed on Earth. Large igneous plateaus occur on both continental and oceanic crust in purely intraplate settings, on present and former plate boundaries (large lithospheric weaknesses), and along the edges of continents, and are interpreted to be the result of plume activity [Coffin and Eldholm,

1994]. The largest terrestrial example of plume-derived igneous plateau growth among topographic basins is Ontong-Java Plateau in the west-central Pacific [Richardson et al.,

2000]. 50

AQUIFER CHARACTERISTICS

The sources of the largest martian outflow channel systems (Circum-Chryse and the proposed Northwestern Slope Valleys) suggest that the infill materials of the gigantic drainage basin may have produced a highly productive aquifer. Large pulses of magmatic activity related to the development of the Tharsis magmatic complex {Dohm et al, 2000a,b] probably resulted in the partial iniSlling of the basin by emplacing stacked sequences of lavas. Evidence of this may be the layered walls of Valles Marineris, which have been interpreted to consist of flood lavas [e.g., Witbecket al, 1991; Lucchitta et al,

1992]. Depending on Noachian climate conditions, these sequences may be interfingered with lacustrine and/or eolian deposits. Furthermore, the basaltic sequences are highly fractured by magmatic and tectonic activities [e.g., Scott and Tanaka, 1986; Tanaka and

Davis, 1988; Tanaka, 1990; Scott and Dohm, 1990a,b; Scott and Dohm, 1997; Anderson et al, 1998, JGR Planets — in press; Dohm et al, 1998; Dohm and Tanaka, 1999; Dohm et al, 2000b; Solomon and Head, 2000]. Highly fractured basalt can have uniisually high values of hydraulic conductivity in terrestrial aquifers. Preliminary analyses of the equations that govern subsurface hydrological behavior imply that this relationship not only remains true for a martian aquifer, but the value of hydrologic conductivity may be amplified due to the lowered frictional resistance afforded by the lesser gravity of Mars

[Dohm et al, 2000c, 2001]. Additionally, any interfingered sedimentary deposits would serve as local reservoirs for subsurface water capable of being tapped by the highly fractured basalts. We estimate the average depth of the Tharsis basin to be between 2 and 7 km (for example. Figures 9-11). The measured area of the basin is approximately 9x10^ km^

(Figures 1, 1, 9-11), and as such, the fill volume for an average depth of 5 km is approximately 4.5x10' km^. If the fill were largely composed of highly fiactured vesicular extrusive basalts, then the porosity and permeability would be very high. On

Earth, the porosity of unfiactured vesicular basalts ranges fi:om roughly 20% for low vesicularity to roughly 75% for highly vesicular scoriaceous basalts [Saar and Manga,

1999]. The porosity and especially the permeability could be even hi^er in the Tharsis region as they are highly fractured from tectonic processes, including faulting. Unlike the highly fiactured surfaces that prevail throughout the Tharsis Magmatic Complex and thus high potential permeabilities, impact crater events may lower porosity and permeability because they will produce fines that infill pore spaces [Mackinnon and

Tanaka, 1989]. When compared to highly cratered surfaces of the southem highlands, however, the proposed basin region (also of the southem highlands) is significantly less cratered.

If the terrestrial porosities are characteristic of the Tharsis basin fill, then the potential volume of water contained in the aquifer would be more than equivalent to the volume of water required to create the putative ocean in the northern plains estimated at

1.4 X 10^ km^ [Head et al, 1999]. For example, if the porosity of basalts was 44%, and the removal of total water from the aqui&r was 60%, the minimum fill volume of the northem plains ocean would be achieved. The inferred porosity is well within the values measured for unfractured basalt flows of moderate porosity [Saar and Manga, 1999]. We realize, however, that it is hard to imagine that 60% of the entire aquifer could have been discharged in one event. In addition, one can only guess at the effectiveness of the proposed aquifer system. Thus, we pose the following considerations. Other hydrogeologic activities associated with catastrophic flooding and related short-lived

(-10"* to 10^ year) episodes of quasi-stable climatic conditions [Baker et al, 1991; Baker et al, 2000] may have also contributed water to the hypothesized oceans by mechanisms such as spring-fed activity along areas of the highland-lowland boundary. Certainly other aquifers may have contributed to the putative northern plains ocean as well, especially following catastrophic flood events and related short-lived climatic perturbations described by Baker et al. In addition, a large quantity of groimd ice (e.g., stagnant ice sheets) may have already been in place in the northern plains during the period(s) of catastrophic flooding IDofim et al, 2000c, JGR Planets in press]. This large quantity of ice would be the likely consequence of an earlier warm and wet phase of Mars, possibly induced by catastrophic flooding [Baker et al., 1991; Baker et al., 2000]. As new floodwaters washed over the northern plains, the additional heat would melt the iqjper layers of ice and the gradients created would allow the melt water to cycle into the hydrologic system. Alternatively, proposed massive debris-flow ocean [Tandka et cd..

Geology in press] and/or a mud ocean [Jons, 1986] require less source water then a putative vast northem ocean [Jons, 1986; Parker et al, 1987, 1993; Head et al, 1999].

In addition, potential paleolakes, which also require less source water, have been m^ped in the northem plains [Scott et al, 1995] 53

DISCUSSION

There are several significant observations/characteristics in the Tharsis and surrounding regions that mi^t be collectively explained by an Early to Middle Noachian drainage basin. These include:

(1) the paucity of Noachian outcrops (Figures 8-11) [e.g., Scott and Tanaka,

1986; Rotto and Tanaka, 1995; Dohm and Tanaka, 1999; Nelson and

Greeley, 1999] and crustal magnetic anomalies in the proposed basin region

[Acuna, et ah, 1999],

(2) a sufiBcient source of water necessary to carve the circum-Chryse outflow

channel systems and the recently proposed structurally-controlled system of

valleys that routed Noachian and Early Hesperian floods from the Arsia Mons

region to the northwest into Amazonis Planitia [Dohm et al, 2000c, 2001]),

(3) the thick sequences of layered materials exposed in the walls of Valles

Marineris by magmatic, tectonic, and erosional activity [Scott and Tanaka,

1986; Witbeck et al, 1991; Lucchitta et al, 1992; Dohm et al, 1998;

McKenzie and Nimmo, 1999; McEwen et al, 1999],

(4) an anomalous concentration of layered ejecta craters in the Solis and

Thaumasia Planae regions [Barlow et al, 2001],

(5) the volatile-rich nature of Valles Marineris [Scott and Tanaka, 1986; Witbeck

et al, 1991; Lucchitta et al, 1992], and

(6) the observed hematite deposits detected in Valles Marineris [Christensen et

al, 1998; Noreen et al, 2000]. The thick sequence of layered materials observed in the canyon walls of Valles

Marineris, for example, may represent the accumulation of volcanic and other types of materials shed into the basin. In addition, the hematite deposits detected in the Valles

Marineris region by the TES instrument may be the result of phreatom^jnatic mixing between magma and groundwater or stirface water within the basin \Noreen et al, 2000;

Chapman and Lucchitta, 2000]. Although the true hydrogeologic complexity of the basin and aquifer system undoubtedly eludes us, the circum-Chiyse catastrophic outflow channels and the NSVs demonstrate the tremendous productivity of this Noachian aquifer. Likewise, the hypothesis that the circum-Chiyse catastrophic outflow channels and Northwestern Slope Valleys resulted from catastrophic flood events is strengthened by the identification the gigantic drainage basin, a potential source for the tremendous amount of water required to sculpt the surface. The source region for catastrophic flooding generally has been thought to be the cratered highlands, which are marked by canyon systems of Valles Marineris, lava flows, and chaotic terrain [Scoff and Tanaka,

1986; Witbeck et al, 1991; Lucchitta et al, 1992]. However, the magnitude and location of the most significant martian flood features in the Tharsis region can be best explained by the existence of a Tharsis basin that facilitated the collection of large founts of volatiles early in martian history [Mastirsfy et al, 1977; Fieri, 1980; Craddock and

Maxwell, 1993; Scott et al, 1995; Can and Chamg, 1997; Tanaka et al, 1998\. In addition, the magmadc-triggered floodwateis that originated from the basin probably represent a major source for bodies of water in the northern plains, including oceans and 55

paleolakes that may have resulted in short-lived climatic perturbations [Baker et al,

1991, 2000].

Magmatic-driven uplifts associated with the development of the Tharsis magmatic

complex subsequently transferred groundwater laterally along structurally controlled

conduits within and outside of the basin. ITiis magmatic-driven activity eventually

forced some of the volatiles out of the basin by catastrophic flooding [Baker et al, 1991,

2000], Such lateral movements of volatiles throughout geologic history would result in

parts of the TMC being volatile-rich and volatile-poor. Impact crater studies have

revealed a region in the Solis and Thaumasia Planae (located south of Valles Marineris in

the south-central part of the basin; Figure 1) where an anomalous concentration of craters displaying multiple-layer ejecta morphologies is correlated with a large area of smaller onset diameters for craters displaying single layer morphologies [Bco'low et al, 2001].

Barlow et al. [2001] proposed that tiie anomalous concentration marks a region that was particularly rich in volatiles (ice near the surface and liquid at greater depths) at the time

of cratering compared to other parts of the Tharsis magmatic complex and the rest of the equatorial region. Thus, tiie impact cratering record may be consistent with a scenario

where magmatic-driven uplifts locally deformed the basin, driving some of the water

from the basin by catastrophic flooding and transferring volatiles from the uplifted

localities into portions of the existing subsurface reservoir. This in turn, may have resulted in a local increase of the hydraulic head. Such a linkage of processes is a predictable consequence of the interaction of magmatic intrusions and the water-bearing 56 strata, which we envisioned to have been a common occurrence throughout the long and complex evolution of the TMC and the enormous Noachian drainage basin.

ALTERNATIVE MODELS

The Basin model is not the only model that offers explanations for some of the significant characteristics of the Tharsis and surrounding regions, including the putative local presence of near-surfece water in the Tharsis region. A model proposed by Clifford

[1995] also predicts near-surface water. Clifford proposed that if the planetary inventory of outgassed H2O exceeded the pore volume of the cryo^here by more than a few percent, a subpermafirost groundwater system of global extent would necessarily result

This interconnected global aquifer allows the downward migration of polar basal melt to result in the upward migration of water at temperate and equatorial latitudes.

Theoretically, Ihis woxjld result in the water that was lost fix)m the crust at these latitudes by sublimation, impact, or catastrophic floods to be replenished. Although Clifford's model can adequately explain the occurrence of near surface water in the Tharsis and surrounding regions, it cannot by itself explain the suite of significant observations and characteristics previously discussed.

The putative presence of local near-surface water in the Tharsis region, as identified by anomalous concentrations of layered ejecta craters in the Solis and

Thaumasia Planae regions, can be explained by other means. A warmer and wetter climate possibly induced fi-om the early (e.g., early to middle Noachian) development of the Tharsis rise [Golombek et al, 2000], for example, may have resulted in the ponding of water into local depressions and infiltration of water into the subsurface as it flowed across the planets' surface. As this water moved through the subsurface, it may have encountered lenses of material that prevented its downward migration, or it may have migrated in the subsurface along the natural gradient formed by the Tharsis rise. In the former case, the aquitards would have led to the creation of perched aquifers. Over time, these perched aquifers may have risen to the near surface or surface. As the martian climate cooled, these near-surface aquifers would have been frozen as interstitial ice. As long as they remained frozen, they would remain trapped there, regardless of the changing gradients caused by the continued development of the Tharsis bulge. This water would only be released by the energy associated with meteoric impact [Newsom,

1980], resulting in the anomalous characteristics of impact craters in Solis and Thaxmaasia

Planae. But this model cannot explain the gigantic quantities of water required to carve the outflow channels.

There is yet another model that might explain the near-surface presence of water among other characteristics. During the early rise of Tharsis [e.g., Banerdt et al, 1992;

Golombek et al., 2000], for example, certain topographic irregularities may have formed due to local and regional volcanic constructs. Further assuming that the topogr^hic irregularities formed early enough in the development of the Tharsis rise that they existed within an early warmer and wetter martian environment [e.g., MasursJ(y et al., 1977;

Fieri, 1980; Golombek et al, 2000], the topographic irregularities would have allowed for the trapping of rainfall and runoff within localized topographic depressions. As the ponded water infiltrated into the groimd, it may have become trapped near the surface as 58 the climate cooled. If it remained frozen, it would remain in situ as Tharsis continued to rise, only to be released upon meteorite impact to produce the anomalous field of layered ejecta craters. A problem with this hypothesis, however, is that geologic mapping has not revealed any Early to Middle Noachian-age volcanic edifices within the basin region, with the exception of incipient Syria Planum [e.g., Dohm and Tanaka, 1999\.

Finally, another model to consider is one presented by Tanaka et al. [2000] in which Syria-Planum-centered fold and thrust structures and a cryosphere perhaps a kilometer or so thick resulted in a huge, largely sealed aquifer system within the

Thaumasia plateau. Although a belt of such structures has been recognized centered about Syria Planum [Schultz and Tanaka, 1994], especially in the Thaumasia region

[Dohm and Tanaka, 1999], a continuous belt within the Thaumasia plateau is not observed in the stratigraphic and paleotectonic records, especially on the northern and western margins of the plateau [Dohm et al, 1999; Anderson et al, JGR Planets - in press]. In addition, numerous basement structures, which are radial about the Tharsis rise, form potential conduits for the migration of ground water away from the rise. Thus, the model presented by Tanaka et al. [2000] is not by itself sufGcient to explain the suite of significant observations/characteristics discussed earlier. Thrust structures, such as those potentially observed in the Coprates rise and Thaumasia highlands regions [Schultz and Tanaka, 1994; Dohm and Tanaka, 1999], however, could have played a role in forming a highly productive Noachian drainage basin/aquifer system. 59

CONCXUSION

Paleotopographic reconstructions based jfrom analyses of synthesized stratigraphic, paleotectonic, erosional, and MOLA-derived topographic map data reveal the potential existence of an enormous drainage basin in the Tharsis region during the

Early to Middle Noachian period. Lavas and sediments partly infilled the basin resulting in a highly productive regional aquifer. The stacked sequences of basin materials were subsequently exposed in the canyon walls of Valles Marineris by magmatic-driven doming, tectonic deformation, and erosion. The basin model may collectively explain:

(1) the paucity of Noachian outcrops and crustal magnetic anomalies in the proposed basin region, (2) the thick sequences of layered materials exposed in the walls of Valles

Marineris, (3) an anomalous concentration of layered ejecta craters in the Solis and

Thaimiasia Planae regions, (4) the volatile-rich nature of Valles Marineris, and (5) the observed hematite deposits detected in Valles Marineris. The basin may also provide a sufficient source of water necessary to carve the circum-Chiyse outflow channel systems and the recently proposed Northwestern Slope Valleys and to form Hesperian and younger putative ocean(s) and (or) paleolakes. 60

REFERENCES

Acuna, M. H., J. E. P. Connemey, N. F. Ness, R. P. Lin, D. Mitchell, C. W. Carlson, J. Mcfadden, K. A. Anderson, H. Reme, C. Mazelle, D. Vignes, P. Wasilewski, and P. Cloutier, Global distribution of crustal magnetization discovered by the Mars Global Surveyor MAG/ER experiment: Science, 284,790-793,1999.

Anderson, R. C., and J. M. Dohm, Magmatic-tectonic evolution of Tharsis, Lunar Planet. Sci. Conf. [CD-ROM], XXXI, abstract 1607,2000.

Anderson, R. C., M. P. Golombek, B. J. Franklin, K. L. Tanaka, J. M. Dohm, J. H. Lias, and B. Peer, Centers of tectonic activity through time for the western hemisphere of Mars, Lunar Planet. Sci. Conf. [CD-ROM], XXK, abstract 1881,1998.

Anderson, R. C., J. M. Dohm, M. P. Golombek, A. Haldemann, B. J. Franklin, K. L. Tanaka, J. Lias, and B. Peer, Significant centers of tectonic activity through time for the western hemisphere of Mars, JGR-Planets, in press.

Anderson, S., and R. E. Grimm, Rift processes at the Valles Marineris, Mars: Constraints from gravity on necking and rate-dependent strength evolution, J. Geophys. Res., 103, 11,113-11,124,1998.

Baker, V.R., Water and the Martian landscape. Nature, 412,228-236,200L

Baker, V. R., and D. J. Milton, Erosion by catastrophic floods on Mars and Earth, Icarus, 23,27-41,1974.

Baker, V. R., R. G. Strom, V. C. Gulick, J. S. Kargel, G. Komatsu, and V. S. Kale, Ancient oceans, ice sheets and the hydrological cycle on Mars, Nature, 352,589-594, 1991.

Baker, V. R., R. G. Strom, J. M. Dohm, V. C. Gulick, J. S. Kargel, G. Komatsu, G. G. On, and J. W. Jr. Rice, Mars' Oceanus Borealis, ancient glaciers, and the MEGAOUTFLO hypothesis, Luncar Planet. Sci. Corf. [CD-ROM], XXXI, abstract 1863,2000. 61

Banerdt, W. B., M. P. Golombek, and K. L. Tanaka, Stress and tectonics on Mars, in Mars, University of Arizona Press, H. H. Kieffer editor, 1992.

Barlow, N. G., Constraints on early events in martian history as derived j&om the cratering record, J. Geophys. Res., 95,14,191-14,201,1990.

Barlow N. G., J. Koroshetz, and J. M. Dohm, Variations in the onset diameter for Martian layered ejecta morphologies and their implications for subsurface volatile reservoirs, Geophys. Res. Lett., 28, No. 16,3095-3099,2001.

Carr, M. H., and F. C. Chuang, Martian drainage densities. J. Geophys. Res., 102,9145- 9152,1997.

Chapman, M. G., H. Masursky, and D. H. Scott, Geologic map of science study area 2, north Kasei Valles, Mars (MTM 25072 quadrangle), USGS Misc. Jnv. Ser. Map I- 2107 (1:500,000), 1991.

Chapman, M. G., and K. L. Tanaka, Geologic maps of the MTM 25062 quadrangle (digital compilation) and the MTM 25067 quadrangle (manual compilation), Kasei Valles region of Mars, USGS Misc. Inv. Ser. Mcp 1-2398 (1:500,000), 1996.

Chapman, M. G., and B. K. Lucchitta, New Mars data suggest Melas Chasma interior deposits are sub-ice volcanoes. Geological Society of America abstract, Reno Nevada, November 9-18, Abstract 50817,2000.

Christensen, P. R. et al.. Results form the Mars Global Surveyor Thermal Emission Spectrometer, Science, 279,1692-1697,1998.

Clifford, S. M., A model for the hydrologic and climate behavior of water on Mars, J. Geophys. Res., 98,10,973-11,016,1993. 62

Coffin, M. F,, and O. Eldholm, Large igneous provinces: crustal structoe, dimensions, and external consequences, Rev. Geophys., 32,11-36, 1994.

Craddock, R. A., and T. A. Maxwell, Geomorphic evolution of the Martian highlands throu^ ancient fluvial processes, J Geophys. Res., 98,3453-3468,1993.

Crown, D. A., K. H. Price, and R. Greeley, Geologic evolution of the east rim of the Hellas basin. Mars, Icarus, 100,1-25,1992.

Crown, D. A., and R. Greeley, Volcanic geology of Hardriaca Patera and the eastern Hellas region of Mars, J. Geophys. Res., 98,3431-3451,1993.

DeHon, R A., Geologic map of the Pompeii quadrangle (MTM 20057), Maja Valles region of Mars, USGSMisc. Inv. Ser. Map 1-2203 (1:500,000), 1992.

Dohm, J. M., and K. L. Tanaka, Geology of the Thaumasia region. Mars: plateau development, valley origins, and magjnatic evolution. Planetary and Space Science, 47,411-431,1999.

Dohm, J. M., R. C. Anderson, and K. L. Tanaka, Digital structural mapping of Mars, Astronojny & Geopf^sics, 39,3.20-3.22,1998.

Dohm, J. M., R. C. Anderson, V. R, Baker, R. G. Strom, G. Komatsu, and T. M. Hare, Pulses of magmatic activity through time: potential triggers for climatic variations on Mars, Lunar Planet. Sci. Corf. [CD-ROM], XXXI, abstract 1632,2000a.

Dohm, J. M., R. C. Anderson, V. R, Baker, C. R. Neal, J. C. Ferris, T. M. Hare, K. L. Tanaka, D. H. Scott, J. A. Skinner, and J. E. Klemaszewski, Evolution of the principal magmatic complex of Mars, Tharsis, Tharsis Workshop abstract. Keystone Colorado, October 16-18,2000b.

Dohm, J. M., R. C. Anderson, V. R, Baker, J. C. Ferris, T. M. Hare, R. G. Strom, L. P. Rudd, J. W. Jr. Rice, and D. H. Scott, System of gigantic valleys northwest of Tharsis 63

Montes, Mars: Latent catastrophic flooding, northwest watershed, and implications for northern plains ocean, Geopf^s. Res. Lett., 27, No. 21,3559-3562,2000c.

Dohm, J. M., V. R. Baker, R. C. Anderson, J. C. Ferris, T. M. Hare, K. L. Tanaka, J. E. Klemaszewski, D. H. Scott, and J. A. Skinner, Martian magmatic-driven hydrothermal Sites: potential sources of energy, water, and life, presented in the Concepts and Approaches for Mars exploration: Lnnar and Planetary Inst., Houston, LPI Contrib. No. 1062,93-94,2000d.

Dohm, J. M., J. C. Ferris, R. C. Anderson, V. R. Baker, Strom, R.G., T. M. Hare, K. L. Tanaka, Barlow, N. G., D. H. Scott, J. A. Skinner, and J. E. Klemaszewski, The Tharsis basin of Mars: A potential source for the outflow channel systems, layered wall deposits of Valles Marineris, & putative oceans or paleolakes, Tharsis Workshop abstract. Keystone Colorado, October 16-18,2000e.

Dohm, J. M., R. C. Anderson, V. R. Baker, J. C. Ferris, L. P. Rudd, T. M. Hare, J. W. Jr. Rice, R.R. Casavant, R. G. Strom, J.R., Zimbelman, and D. H. Scott, in press. Latent activity for western Tharsis, Mars: significant flood record exposed, J. Geophys. Res., 106,12,301-12,314,2001.

Frey, H., Thaumasia: A Fossilized Early Forming Tharsis Uplift, J. Geophys. Res., 84, 1009-1023,1979.

Frey, H., T. D. Grant T.D., Resurfacing history of Tempe Terra and surroundings, J. Geophys. Res., 95,14,249-14,263,1990.

Golombek, M. P., Tharsis tectonics, Tharsis Workshop abstract. Keystone Colorado, October 16-18,2000.

Golombek, M. P., W. B. Banerdt, and R. J. Phillips, Tectonics of Mars, Geological Society of America abstract, Reno Nevada, November 9-18, Abstract 51301,2000.

Greeley, R, and J. E. Guest, Geologic m^ of the eastern equatorial region of Mars, USGSMisc. Inv. Ser. Mapl-1802B (1:15,000,000), 1987. 64

Gulick V. C., Magmatic intrusions and hydrothermal systems; Implications for the formation of small Martian valleys. PhD, Thesis, Univ. of Arizona, Tucson, Arizona, 1993.

Head HI, J. W., H. Hiesinger, M. A. Ivanov, M. A. Kreslavsky, S. Pratt, and B. J. Thompson, Possible ancient oceans on Mars: evidence from Mars Orbiter Laser Altimeter data. Science, 286,2134-2137,1999.

Head in, J. W., B. M. Webb, B. E. Kortz, and S. Pratt, Syria Planum, Mars: A major volcanic construct in the early history of Tharsis, Tharsis Workshop abstract. Keystone Colorado, October 16-18,2000.

Jons, H. P., Arcuate ground imdulations, gelifluxion-like features and "front tori" in the northern lowlands on Mars, what do they indicate?, (abstract). Lunar. Planet. Sci. Cow/, XVII, 404-405,1986.

Lucchitta, B. K., A. S. McEwen, G. D. Clow, P. E. Geissler, R. B. Singer, R, A. Schultz, and S. W. Squyres, The canyon system on Mars, in Mars, pp. 453-492, Univ. of Ariz. Press, Tucson, 1992.

MacKinnon, D. J., and K. L. Tanaka, The impacted martian crust: structure, hydrology, and some geologic implications. J. Geophys. Res., 94,17,359-17,370,1989.

Masursky, H., J. M. Boyce, A. J. Dial, G. G. Schaber, and M. E. Strobell, Formation of Martian channels,./ Geophys. Res., 82,4016-4038,1977.

McEwen, A. S., M. C. Malin, M. H. Carr, and W. K. Hartmann, Voluminous volcanism on early Mars revealed in Valles Marineris, Natwre, 397,584-586,1999.

McKenzie, D., and F. Nimmo, The generation of Martian floods by the melting of groimd ice above dykes, Nature, 397,231-233,1999. 65

Milton, D. J., Geologic map of the Lunae Pains quadrangle of Mars: USGS Misc. Im. Ser. Map 1-894 (1:5,000,000), 1974.

Morris, E. C., and K. L. Tanaka, Geologic map of the Olympus Mons region of Mars, USGS Misc. Inv. Ser. Map 1-2327,1994.

Mouginis-Mark, P. J., Volcano/ground ice interactions in Elysium Planitia, Mars, Icarus, 64,265-284,1985.

Nelson, D. M., and R. Greeley, Geology of Xanthe Terra outflow channels and the Mars Pathfinder landing site, J. Geophys. Res., 104,8653-8669, 1999.

Newsom, H. E., Hydrothennal alteration of impact melt sheets with implications for Mars; Icarus, 44,207-216,1980.

Noreen E., K. L. Tanaka, M. Chapman, G. Haxel, and L. Gaddis, Late stage magmatic- hydrothermal activity in eastem Tharsis?, Tharsis Workshop abstract. Keystone Colorado, October 16-18,2000.

Parker, T. J,, D. M. Schneeberger, D. C. Pieri, and R. S. Saunders, Geomoiphic evidence for ancient seas on Mars, in Symposium on Mars: Evolution of its Climate and Atmosphere, LPITech Kept. 87-01,96-98,1987.

Parker, T. J., D. S. Gorsline, R. S. Saunders, D. C. Pieri, and D. M. Schneeberger, Coastal geomoiphology of the martian northern plains, J. Geophys. Res., 98, 11,061-11,078, 1993.

Pieri, D. C., Geomorphology of Martian valleys, in Advance in Planetary Geology: National Aeronautics and Space Administration Technical Memorandum 81979, 1- 160,1980.

Plescia, J. B., L. E. Roth, and R. S. Saimders, Tectonic features of southeast Tharsis. In Reports of Planetary Geology Proff-am, NASA Tech. Memo. 81776,68-70. 1980. iI

66

Plescia, J. B., and R. S. Saunders, Tectonic history of the Tharsis region. Mars, J. Geophys. Res., 87,9775-9791,1982.

Rice, J. R., Jr., and R. A. DeHon, Geologic map of the Darvel Quadrangle (MTM 20052), Maja Valles region of Mars, USGS Misc. Jnv. Ser. Map 1-2432 (1:500,000), 1996.

Richardson, W. P., E. A. Okal, S. Van der Lee, 2000, Rayleigh-wave tomography of the Ontong-Java Plateau: Physics of the Earth and Planetary Interiors, 118,29-51,2000.

Robinson, M. S., P. J. Mouginis-Mark, J. R. Zimbehnan, S. S. C., Wu, K. K. Ablin, and A. E. Howington-Kraus, Chronology, eruption duration, and atmospheric contribution of the Martian volcano Apollinaris Patera. Icarus, 104,301-323, 1993.

Rotto, S. L., and K. L. Tanaka, Geologic/geomorphologic map of the Chryse Planitia region of Mars, USGS Misc. Inv. Ser. Map 1-2441 (1:5,000,000), 1995.

Saar, M. O., and M. Manga, Permeability-porosity relationship in vesicular basalts, Geophys. Res. Lett., 26,111-114,1999.

Schubert, G., S. C. Solomon, D, L. Turcotte, M. J. Drake, and N. H. Sleep, Origin and thermal evolution of Mars: in Mars, University of Arizona Press, Tucson, pp. 147- 183,1992.

Schultz, R. A., and K. L. Tanaka, Lithospheric-scale buckling and thrust structures on Mars: The Coprates rise and south Tharsis ridge belt, J. Geophys. Res., 99, 8371- 8385,1994.

Scott, D. H., Greologic map of the MTM 25057 and 25052 quadrangles, Kasei Valles region of Mars, USGS Misc. Inv. Ser. I-Map 2208 (1:500,000), 1993.

Scott, D. H,, and J. M. Dohm, Faults and ridges: Historical development in Tempe Terra and Ulysses Patera regions of Mars. In Proc. of the 20^ Lunar Planet. Sci. Corf., 503-513,1990a. 67

Scott, D. H., and J. M. Dohm, Tectonic setting of Martian volcanoes and deep-seated intrusives. In MEVTV Workshop "Evolution on Magma Bodies on Mars", 39-40, 1990b.

Scott, D. H., and K. L. Tanaka, Geologic map of the westem equatorial region of Mars, USGSMisc. Inv. Ser. MapI-1802-A (1:15,000,000), 1986.

Scott, D. H. and J. R. Zimbelman, Geologic map of Arsia Mons volcano. Mars, USGS Misc. Inv. Ser. Map 1-2480 (1:1,000,000), 1995.

Scott, D. H., and J. M. Dohm, 1997, Mars: structural geology and tectonics: In Encyclopedia of Planetary Sciences, eds. J.H. Shirley, and R.W. Fairbridge: Van Nostrand Reinhold, New York, pp. 990.

Scott, D. H., J. M. Dohm, and D. J. Applebee, Geologic map of science study area 8, ApoUinaris Patera region of Mars, USGS Misc. Inv. Ser. Map 1-2351 (1:500,000), 1993.

Scott, D. H., J. M. Dohm, and J. W. Jr. Rice, Map of Mars showing channels and possible paleolake basins, USGSMisc. Inv. Ser. Mcq> 1-2461 (1:30,000,000), 1995.

Scott, D. H., J. M. Dohm, and J. R. Zimbelman, Geologic map of Pavonis Mons volcano. Mars, USGSMisc. Inv. Ser. Map 1-2561 (1:1,000,000), 1998.

Sleep, N. H., Martian plate tectonics. J. Geophys. Res., 99,5639-5655,1994.

Solomon, S. C., and J. W. Head, Evolution of the Tharsis Province of Mars: The Importance of Heterogeneous Lithospheric Thickness and Volcanic Construction, J. Geopf^s. Res., 87,9755-9774,1982. 68

Solomon, S. C., and J. W. Head, The nature and evolution of Tharsis: global context and new insights from MGS observations, Tharsis Workshop abstract. Keystone Colorado, October 16-18,2000.

Tanaka, K. L., The stratigraphy of Mars, Proc. of the 17'^ Lunar Planet. Sci. Conf, in J. Geophys. Res.,pt. 1,91, E139-E158, 1986.

Tanaka, K. L., Tectonic history of the Alba Patera-Ceraunius Fossae region of Mars. In Proc. of the 2(f^ Lunar Planet. Sci. Conf, 515-523, 1990.

Tanaka, K. L., and P. A. Davis, Tectonic history of the Syria Planimi province of Mars, J. Geophys. Res., 93,14,893-14,917,1988.

Tanaka, K. L., J. M. Dohm, J. H. Lias, and T. M. Hare, Erosional valleys in the Thaumasia region of Mars: Hydrotheimal and seismic origins; J. Geophys. Res., 103, 31,407-31,419,1998.

Tanaka, K. L., J. M. Dohm, J. C. Ferris, and R. C. Anderson, Erosional mechanism for the development of Valles Marineris, Tharsis Workshop abstract. Keystone Colorado, October 16-18,2000.

Tanaka, K. L., Banerdt, W. B., Kargel, J. S., and N. HofBnan, Huge, COa-charged debris- flow deposit and tectonic sagging in the northern plains of Mars, submitted to Geology.

Wise, D. U., M. P. Golombek, and G. E. McGill, Tharsis province of Mars: Geologic sequence, geometry, and a deformation mechanism, Icarus, 38, A56A72,1979.

Witbeck, N. E., K. L. Tanaka, and D. H. Scott, Geologic map of the Valles Marineris region. Mars (east half and west half), USGS Misc. Inv. Ser. Map 1-2010 (1:2,000,000), 1991. 69

Zimbelman, J. R., S. E. H. Sakimoto, and H. Frey, Evidence for a fluvial contribution to the complex story of the Medusae Fossae Formation on Mars, Geol Soc. Am. An Meeting, abstract 50423,2000. Stage Period

inkrsi

Figure 1. MOLA shaded relief map of the western hemisphere of Mars (courtesy of the

MOLA Science Team). Warmer colors indicate higher elevations, while cooler colors indicate lower elevations. Shown are the major geologic features of the Tharsis

Magmatic Complex (TMC) and their stages of development (stage assignments 1-5 of 71

Dohm and Tanaka [1999] correlated with the time-stratigraphic age assignment—

Noachian, Hesperian, and Amazonian periods of Tanaka [1986\), based on numerous detailed geologic investigations [e.g., Milton, 1974; Baker and Milton, 1974; Plescia et al, 1980; Plescia and Saunders, 1982; Scott and Tanaka, 1986; Tanaka, 1986; Tanaka and Davis, 1988; Frey and Grant, 1990; Tanaka, 1990; Scott and Dohm 1990a,b; Baker et al., 1991; Witbeck et al, 1991; Chapman et al, 1991; Morris et al, 1991; Lucchitta et al, 1992; DeHon, 1992; Scott, 1993; Gulick, 1993; Morris and Tanaka, 1994; Schultz and Tanaka, 1994; Rotto and Tanaka, 1995; Scott and Zimhelman, 1995; Scott et al,

1995; Rice and DeHon, 1996; Chapman and Tanaka, 1996; Scott and Dohm, 1997; Scott et al, 1998; Dohm et al, 1998; Anderson et al, 1998; Dohm and Tanaka, 1999; Nelson and Greeley, 1999; McKenzie and Nimmo, 1999; Anderson and Dohm, 2000; Dohm et al, 2000b; Chapman and Lucchitta, 2000; Head et al, 2000; Baker et al, 2001; Dohm et al, 2001; Anderson et al, JGR Planets - in press}. Also shown are Solis and Thaumasia

Planae (S.P. and T.P., respectively), approximated margins of the TMC (black dashed outline), northwestem slope valleys (NSVs), and the Noachian drainage basin (blue dashed outline). I J

I

72

FEATURES OF TMC

<2 u>

O) at

Figure 2. Chart comparing the stages of geologic activity in the Tharsis Magmatic Complex region with major geologic features, including the Noachian drainage basin (correlates to Figure 1), Size of solid areas roughly proportional to degree of exposed deformation. Violet—centers of tectonic activity inteipreted to be the result of magmatic-driven uplift and local volcanism, dike emplacement, and hydrothermal activity; orange—^moimtain building; blue—^water; red—primarily emplacement of shield-forming and lava-field-forming lavas. Note that the commencement and/or end of activity of the components of the complex are not absolutely constrained and that features such as the shield volcanoes of Tharsis Montes and Olympus Mons could be presently active. 73

'Uranii

I" -'cir^^^

Syria Planum^ 74

Figure 3. 3-D portrayal of the major geologic features of the Tharsis Magmatic Complex using MOLA data (Stage 5—Amazonian [Dohm and Tanaka, 1999]). plai^ui

;^'n1s 'fflowWajt#

V&insr " ,^Lrgyre i^teact

I 76

Figure 4. 3-D portrayal of the major geologic features of the Tharsis Magmatic Complex during the Stage 4 (Late Hesperian-Early Amazonian p)ohm and Tanaka, 1999]). When compared to Stage 5 (Figure 3), significant highlights include: (1) less developed shield volcanoes, including Olympus Mons, Tharsis Montes, and Alba Patera and their associated lava flows, (2) recession of the highland-lowland boundary, (3) less modified

Thaumasia plateau (cessation of plateau uplift is mapped as Early Hesperian [Dohm and

Tanaka, 1999], and (4) less infilled and more defined basins (e.g., Chryse and Argyre), canyons (e.g., Valles Marineris), as weU as deeper outflow channels. •5^ TaSipe Terra' plateau":"'-

circumjChryse : " :Pavonis ifflow channel system§5

Th^masl plateau Uzboj, V^llis: i - \A^ego- Figure 5. 3-D portrayal of the major geologic features of the Tharsis M^matic Complex during the Stage 3 (Early Hesperian [Dohm and Tanaka, 1999]). When compared to

Stage 4 (Figure 4), significant highlights include; (1) absence of Olympus Mons and

Tharsis Montes shield volcanoes, including associated lava flows, (2) incipient Pavonis rise, (3) less developed Alba Patera and associated lava flows, (3) more prominent Syria

Planimi, Ceraunius Fossae, Uranius Fossae, central Valles Marineris, Arsia, Claritas, and

Warrego rises, and the NSVs (4) recession of the highland-lowland boimdary, (5) less modified Thaumasia (cessation of plateau uplift is mapped as Early Hesperian [Dohm and Tanaka, 1999] and Tempe plateaus, and (6) less infilled and more defined basins

(e.g., Solis, Chryse, and Argyre) and smaller and less defined outflow channels. 79

IUS Fossae ateau-^ Chryse impact ?

raniusdboss

, arcum/Jhryse

'^trai \&Qes & Marinerisnse

•4 Thaumasi Oantas plateau Uzboi

•• Argyre •^7 rise impact' ,# 80

Figure 6. 3-D portrayal of the major geologic features of the Tharsis Magmatic Complex during the Stage 2 OLate Noachian-Early Hesperian [Dohm and Tanaka, 1999]). When compared to Stage 3 (Figure 5), significant highlights include: (1) absence of Alba Patera and associated lava flows, (2) less prominent Warrego and Arsia rises, (3) incipient trough development at Valles Marineris, (3) more prominent mountain ranges of

Coprates rise and Thaumasia highlands, Tempe Terra and Tbaumasia igneous plateaus,

Syria Planum, Ceraunius Fossae and Claritas rises, (3) recession of the highland-lowland boundary, (4) less modified Thaumasia and Tempe plateaus, and (6) less infilled and more defined basins (e.g., Solis, Chryse, and Argyre), the NSVs, and Uzboi Vallis; smaller and less defined outflow channels. 81

UraniuSx,^ nse?^. """I

lianum Figure 7. 3-D portrayal of the major geologic features of the Tharsis Magmatic Complex

{Solomon and Head, 2000; Dohm et al., 2000b; Anderson et al., JGR-Planets in press] during the Early to Middle Noachian Period (Part of Stage 1 [Dohm and Tanaka, 1999\), including the drainage basin. When compared to Stage 2 (Figure 6), significant highlights include: (1) drainage basin, (2) narrower and sharper mountain ranges of

Coprates rise, Thaumasia highlands, and multi-ringed structures of the Argyre impact basin (the Charitum and Nereidum Montes), (3) incipient Ceraunius Fossae, Claritas, and

Arsia rises, Syria Planimi, and Tempe plateau, (4) absence of highland-lowland boundary, Valles Marineris, circimi-Chryse outflow channel systems, and Thaumasia plateau, (5) more distinct Argyre and putative Chryse impact basins, and (6) less distinct

Uzboi Vallis. 83

Plyum^ Amm ..,0^ I -— • -1 • r • / I .•

N^ical cxasgcraiifn 150 X dl d2 d3 d4 d5 a' 7147 i ^ TO! A 1 !r \ ^ 45» -1 TO! I 2 1 V / \ Nt:lh Shalbatana ^^pJr \Simud 1 \ Tiu -2 -2151 ! • Np4i >^y Ares: U2 of pLertial msltsials (MF, Kas«| N?M -5S06 i^|wvo Tnaf5is«reiaicdvolcaaw, flux'lal. eoban, glaciar?),- •GTU) ! PBS PBS i PBS i i PBS 2i2<»;KU JUi5«U <.QG7l2\i 7JS5SW SJlUtti Distance (m) 84

Figures 8 (A) MOLA shaded relief map showing features of interest, including NSVs,

Tharsis rise, Circum-Chryse outflow channel systems, and centers of tectonic activity

(Anderson et al., 1998; Dohm et al., 1998) interpreted to represent magmatic-driven doming events (UD - Ulysses, ASD - Ascaeus-south; ASWD - Arsia-southwest; CVD - central Valles; SD - Syria), (B) part of the Geologic map of the western equatorial region of Mars (representative map units are shown-Scott and Tanaka, 1986), (C) generalized geologic cross section (a-a'—transect of A and B; PBS - potential location of basement stractures; TMl- Tharsis Montes lavas; MF - Medusae Fossae materials) [Figure 2 of

Dohm et al, JGR-Planets in press]. I I

85

A Vertical exaggeration 75 X ccntral Vallcs Marinoris rise. Thaumaisa Highalnds

impact rim? 2680 Dislancc (km)

1000 0 1000 KHomotera

Figure 9. (A) Present-day MOLA profile (Transect A-A') across the west-central part of the Thaumasia highlands, the southeast part of the Noachian drainage basin (queried blue line represents uncertain basin extent), including the central Valles Marineris rise (center of tectonic activity, interpreted to be the result of m^matic-driven uplift [Dohm et al., 1998; Anderson et al., 1998, JGR-Planets in press; Dohm et al., 2000b]), and materials of inferred rim of Chryse impact basin, (B) MOLA shaded relief map showing features of interest, including the approximated boundary of the Noachian drainage basin (dashed blue line) and the central Valles Maiineris rise, and (C) part of the geologic map of the western equatorial region of Mars (representative map units are shown—[Scott and Tanaka,1986]). 86

Profile 6 Vertical exaggeration 75 X Warrogo Rise (MTLC)

Thaumasia Wostvallcs g-436 Highlands Marineris Rise (MTLC) Tempo Terra 03-3136 Plateau

-5836 1022 2044 3066 4088 5110 6131 Distance (km)

V /f

1000 1000 Kilometers

Figure 10. (A) Present-day MOLA profile (Transect A-A') across the Warrego rise (center of tectonic activity, interpreted to be the result of magmatic-driven uplift [Anderson et al., 1998, JGR-Planets in press; Dohm et al., 1998; Dohm and Tanaka, 1999; Dohm et al., 2000b]), west-central part of Thaumasia highlands, central part of the Noachian drainage basin (queried blue line represents uncertain basin extent), including west-central Valles Marineris rise (center of tectonic activity, interpreted to be the result of magmatic-driven uplift [Anderson et al., 1998, JGR-Planets in press; Dohm et al., 1998; Dohm and Tanaka, 1999; Dohm et al., 2000b]), and Tempe Terra plateau, (B) MOLA shaded relief map showing features of interest, including the approximated boxmdary of the Noachian drainage basin (dashed blue line) and the west-central Valles Marineris rise, and (C) part of the geologic map of the western equatorial region of Mars (representative map units are shown—[Scott and Tanaka, 1986]). 87

A Vertical exaggeration 75 X

16789

Mrsia' Mons

Syria Planum

2629 Distance (km)

Figure 11. (A) Present-day MOLA profile (Transect A-A') across the Coprates rise [Schultz and Tanaka, 1994; Dohm and Tanaka, 1999], southern part of the Noachian drainage basin (queried blue line represents uncertain basin extent), Syria Planum (center of tectonic activity, interpreted to be the result of magmatic-driven uplift and volcanism [Tanaka and Davis, 1988; Anderson et al., 1998, JGR-Planets in press; Dohm et al., 1998; Dohm and Tanaka, 1999; Dohm et al., 2000b]), Arsia rise (center of tectonic activity, interpreted to be the result of magmatic-driven uplift and volcanism [Anderson et al., 1998, JGR-Planets in press; Dohm et al., 2000b], and Arsia Mons [Scott and Tanaka, 1986; Scott and Zimbelman, 1995], (B) MOLA shaded relief map showing features of interest, including the approximated boundary of the Noachian drainage basin (dashed blue line), and (C) part of the geologic map of the western equatorial region of Mars (representative map units are shown—[Scott and Tanaka, 1986]). 88

APPENDIX B

LATENT OUTFLOW ACTIVITY FOR WESTERN THARSIS, MARS: SIGNIFICANT FLOOD RECORD EXPOSED.

J.M. Dohm, R.C. Anderson, V.R. Baker, J.C. Ferris, L.P. Rudd, T.M. Hare, J.W. Rice, Jr., R.R. Casavant, R.G. Strom, J.R. Zimbelman, D.H. Scott, and J.A. Skinner, Jr.

Manuscript published in the Journal of Geophysical Research, Volume 106, Pages 12,301-12,314, June 25^ 2001. 89

FIRST AUTHOR PERMISSION

I hereby grant permission to Mr. Justin C. Ferris to include the following manuscript in this dissertation submitted to the faculty of the Department of Hydrology and Water Resources and the Graduate College of the University of Arizona.

Manuscript Title; Latent outflow activity for western Tharsis, Mars: Significant flood record exposed

Date

James M. Dohm 90

LATENT OUTFLOW ACTIVITY FOR WESTERN THARSIS, MARS:

SIGNIFICANT FLOOD RECORD EXPOSED.

J.M. Dohm', R.C. Anderson^ V.R. Baker'"^, J.C. Ferris', L.P. Rudd', T.M. Hare\ J.W.

Rice, Jr.^, R.R. Casavant', R.G. Strom^ J.R. Zimbelman^ D,H. Scott®, and J.A. Skinner,

Jr.'* 'Department of Hydrology and Water Resources, University of Arizona, Tucson,

AZ, 85721, [email protected]: Jet Propulsion Laboratory, Pasadena, CA; Lunar and Planetary Laboratory, University of Arizona, Tucson, AZ; ''U.S. Geological Survey,

Flagstaff, AZ; ^Smithsonian Institution, Washington, DC; ^vate.

ABSTRACT

Observations permitted by the newly acquired Mars Orbiter Laser Altimeter

(MOLA) data have revealed a system of gigantic valleys northwest of the huge martian shield volcano, Arsia Mons, in the western hemisphere of Mars (northwestern slope valleys—NSVs). These features, which generally correspond spatially to gravity lows, are obscured by veneers of materials including volcanic lava flows, air fall deposits, and eolian materials. Geologic investigations of the Tharsis region suggest that the system of gigantic valleys pre-dates the construction of Arsia Mons and its extensive associated lava flows of mainly Late Hesperian and Amazonian age and coincides stratigraphically with the early development of the outflow channels that debouch into Chiyse Planitia.

Similar to the previously identified outflow channels, which issued tremendous volvmies of water into topographic lows such as Chiyse Planitia, the NSVs potentially represent 91 flooding of immense magnitude, and as such, a source of water for a northern plains ocean.

INTRODUCTION

The images of Mars acquired by the Mariner 9 and Viking orbiters during the 1970s revealed geologic terrains that display assemblages of geomorphic features characteristic of flood-carved surfaces on Earth such as the Channeled Scabland of the western United

States [Bretz, 1969; Baker and Milton, 191A-, Baker, 1978, 19^2-, Mars Channel Working

Group, 1983]. Assemblages of relict landfonns, which include anastomosing channel patterns, streamlined islands, longitudinal grooves, scour depressions, depositional bars, inner channels, and channel sources, strongly indicate ancient catastrophic flood events on Mars. Although smaller outflow channels are recorded elsewhere on Mars including

Mangala [Scott and Tanaka, 1986; Chapman and Tanaka, 1993; Zimbelman et ah, 1994;

Craddock and Greeley, 1994], Ma'adim [Greeley and Guest, 1987; Scott et al., 1993;

Scott and Chapman, 1995], Dao [Greeley and Guest, 1987; Crown et ah, 1992; Crown and Greeley, 1993; Scott et al, \995-, Price, 1998], Harmahkis 1987;

Crown et al., 1992; Crown and Greeley, 1993; Scott et al, 1995; Price, 1998], ReuU

[Greeley and Guest, 1987; Crown et al, 1992; Crown and Greeley, 1993; Scott et al,

1995; Price, 1998], and Marte [Plescia, 1990; Scott et al, 1995] Valles, the large outflow channels, which debouch into Chiyse Planitia (circum-Chryse outflow channel systems;

Figure 1) and generally source near broken surfaces marked by the Valles Marineris canyon system and by fiactures, hiUs, and mesas forming chaotic terrain, have received 92 the greatest attention [e.g., Milton, 1974; Baker and Milton, 1974; Scott and Tanaka,

1986; Chapman et al., 1991; DeHon, 1992; Scott, 1993; Rotto and Tanaka, 1995; Scott et al, 1995; Rice and DeHon, 1996; Chapman and Tanaka, 1996; Nelson and Greeley,

1999]. Geologic mapping indicates that the outflow systems may have been carved by several flood episodes over protracted time intervals. For example, Kasei Valles is one of the largest previously identified systems of outflow channels on Mars and a major contributor of water to Chryse Planitia. It formed by multiple flood events during the

Hesperian and Early Amazonian [e.g., Scott, 1993]. Another example is Mangala Valles where at least two episodes of flooding have been recorded [Chapman and Tanaka, 1993;

Zimbelman et al, Craddock and Greeley, 1994].

One commonly held explanation for the origin of the outflow channels involves episodic mj^matic-driven processes [e.g., Baker et al., 1991; Dohm et al., 2000], perhaps aided by gas hydrate dissociation [Max and ClifFor4 2000; Kargel et al., 2000; Komatsu et al., 2000], that explosively expel ground water and other materials (e.g., water-rich debris) fix)m an ice-enriched crust [e.g.. Baker and Milton, 1974; Masursky et al, 1977;

Carr, 1979; Nrmimedal and Prior, 1981; Lucchitta, 1982; MacKinnon and Tanaka, 1989;

Baker et al., 1991]. The expelled materials are catastrophically discharged down gradient towards the northern plains, sculpting the martian surface and entraining boulders, rock, and sediment during passage. The sediment-charged water eventually enters the northern plains contributing to the formation of large bodies of water and sediment, including various hypothesized oceans, seas, lakes, and ice sheets [e.g., McGill, 1985; Jons, 1986;

Lucchitta et cd., Parker et al, 1987, 1993; Baker et al, 1991; Scott et al, 1991a,b; Chapman, 1994; Scott et aL, 1995; Scott and Chapman, 1995; Kargel et al, 1995; Head et al, 1999]. The "MEGAOUTFLO" hypothesis [Baker et al, 2000] first proposed by

Baker et al. [1991] genetically links such activity to relatively short-lived climatic perturbations from a common cold and dry Mars (e.g., short-term hydrological cycles), which includes enhanced erosion by precipitation, landslides/mass wasting (both land and submarine), glacial activity, and the presence of large standing bodies of water such as the Amazonian-aged Oceanus Borealis.

Here, we discuss a system of immense valleys (northwestem slope valleys—NSVs;

Figure 1) newly identified using MOLA data. The system of valleys has largely gone unnoticed since it was first imaged by the Mariner and Viking orbiters because it is extensively veneered by Late Hesperian and younger materials, including lava flows and ash deposits of the Tharsis Montes shield volcanoes and middle Amazonian and younger putative ash flow deposits of the Medusae Fossae Formation [Malin, 1979; Scott and

Tanaka, 1986; Greeley and Guest, 1987]. Though several modes of formation caimot be ruled out for the NSVs (see Section 4), geologic and paleohydrological investigations and topographic and geophysical analyses using MOLA and gravity data indicate that the

NSVs likely represent structurally-controlled valleys that routed catastrophic floods.

Such potential latent outflow activity of western Tharsis may have contributed tremendous volumes of water to the northem plains, an explanation consistent with the putative existence of large bodies of water in the northem plains and the

"MEGAOUTFLO" hypothesis of Baker et al. [1991; 2000]. 94

GEOLOGIC AND PHYSIOGRAPHIC SETTING

A conspicuous system of gigantic, northwest-trending valleys as much as 200-km wide, referred to as the NSVs, are partly defined by enormous elongated promontories and are located along the margin of the northem lowland plains and the southern highlands, west of the huge northeast-trending chain of Tharsis Montes shield volcanoes

(Figure 1). The NSVs occur within a large topographic depression (Figure 2) interpreted to be the result of long-term crustal response to loading by Tharsis [Phillips et al, 2000] and generally correspond to gravity lows. The northem and southern provinces represent a physiographic and geologic dichotomy of the martian crust. Several hypothesis have been advanced for the origin of the dichotomy including mantle convection associated with core formation [Wise et al, 1979], a colossal impact that formed a basin in the north polar region [Wilhelms and Sqiiyres, 1984], extensive southward erosional retreat of the heavily cratered highland plateau [Scott, 1978; Hiller, 1979], and plate tectonics [Sleep,

1994]. Regardless of origin, the boundary clearly separates relatively young, imcratered materials of the lowlands firom the highly cratered, ancient highland rock assemblages.

The lowland materials are interpreted to have been emplaced by both subaerial and subaqueous processes, which include eolian, volcanic, fluvial, mass-wasting, lacustrine, and marine [e.g., Scott and Tanaka, 1986; Tanaka, 1986; Greeley and Guest, 1987;

Parker et al, 1993; Scott et al, 1995;]. The ancient highlands consist of a melange of rocks that are interpreted to mainly include lava flows, impact breccias, and eolian, fluvial, and colluvial deposits [e.g., Scott and Tanaka, 1986; Tanaka, 1986; Greeley and

Guest, 1987; Dohm and Tanaka, 1999], although glacial deposits have also been proposed [e.g., Kargel et al, 1995]. Amazonis Planitia, located to the north of and down gradient from the NSVs, is the site of a hypothesized ocean [e.g., Parker et al., 1993] and

(or) paleolake [e.g., Scott et al., 1995]. Heavily cratered plains materials crop out up gradient and to the south of the NSVs [e.g., Scott and Tanaka, 1986]. A more abrupt change in slope occurs to the east and northeast of the NSVs as a result of the prominent shield volcanoes and associated lava flows of Tharsis Montes [Scott and Tanaka, 1986;

Scott and Zimbelman, 1995; Scott et al., 1998] and Olympus Mons [Scott and Tanaka,

1986; Morris and Tanaka, 1994], respectively (Figures 1 and 2).

Arsia Mons is the southernmost shield volcano of the northeast-trending volcanic chain of Tharsis Montes, located between the NSVs and Syria Planum (Figure 1). Arsia exhibits a summit caldera, reaches a height of greater than 17 km, and displays a basal diameter greater than 300 km. Some of the youngest flows of Arsia overflowed the northeast rim of the caldera while other young flows were extruded from fissures and collapse pits on the northeast and southwest sides of the volcano along the trend of a regional fault zone [Carr et al, 1977; Scott and Tanaka, 1986; Scott and Zimbelman,

1995]. The lower northwestern flank of Arsia Mons is covered by fan-shaped deposits composed of three distinct facies [Scott and Zimbelman, 1995] similar to smaller occurrences on the northwest flanks of Pavonis [Scott et al., 1998] and Ascraeus [Scott and Tanaka, 1986] Montes; the facies may indicate mass movement such as landsliding and debris avalanching, whereas others are characteristic of facies suggesting ash-flow or glacial-periglacial processes. Lava flows emanating from Arsia Mons extend to the northwest at least as far as the newly identified NSVs, as evidenced by MOLA data that 96

shows lava flows traceable j&om within the NSVs to near the western base of Arsia Mons

[Zimbelman et ah, 2000].

Syria Planum including Noctis Fossae, located to the east of Arsia Mons, forms

the western margin of a complex system of canyons, Valles Marineris (Figure 1), Syria

is a site of long-lived magmatic/tectonic activity including domal uplift and associated

radial and concentric faulting and volcanism [Tanaka and Davis, 1988; Anderson et al,

1997; 1998, 1999, 2000; Dohm and Tanaka, 1999]. Valles Marineris appears to have

developed, in large part, along fault systems associated with the early development of the

Tharsis rise [e.g., P/e5cia an£? Sflzojders, 1982]. Additionally, rifting, magma withdrawal, and tension fiacturing have been proposed as possible processes involved in the initiation

and development of the canyons. Lucchitta et al. [1992] noted that the depth of the large

troughs may have been caused by (1) coUapse of near-surface materials due to

withdrawal of imderlying material or opening of tension fiactures at depth; (2)

development of keystone grabens at the crest of a bulge; or (3) failure and subsequent

drifting of plates. Lucchitta [1987] also recognized that many of the valley faults associated with Valles Marineris may have been associated with volcanic activity.

Extending fix>m the northern and eastern margins of Valles Marineris into Chryse Flanitia

are well-documented circvun-Chryse outflow channel systems that cut heavily and

moderately cratered surfaces of the southern cratered highlands (Figures 1 and 2). 97

STRATIGRAPHY

The stratigraphic positions of rock units in the region of interest (The NSVs,

Tharsis, Valles Marineris, and circum-Chryse outflow channel systems; Figure 1) have been established regionally and planetwide by previous mapping investigations including impact crater counts [e.g., Scott and Tanaka, 1986; Tanaka, 1986; Scott and Chapman,

1991c; Morris et al, 1991; Chapman et ah, 1991; Witbeck et ah, 1991; DeHon, 1992;

Scott, 1993; Chapman and Tanaka, 1993; Craddock and Greeley, 1994; Zimbelman et al., 1994; Morris and Tanaka, 1994; Scott and Zimbelman, 1995; Rotto and Tanaka,

1995; Scott et al, 1995; Chapman and Tanaka, 1996; Rice and DeHon, 1996; Scott et al.,

1998; Schultz, 1998; Dohm and Tanaka, 1999; Nelson and Greeley, 1999]. The stratigraphic history of the region of interest is summarized below and portrayed in cross section (Figure 2C).

Noachian System, The Noachian System consists of ancient crustal rocks formed during the period of late heavy bombardment that produced a high density of impact craters including enormous basins. Impact cratering, along with water and wind erosion, tectonic deformation, and magmatic-related resurfacing have modiiSed or destroyed many of the primary morphologic features of Noachian units, including a substantial part of the ancient crater population. Thus, Noachian surfaces are usually characterized by impact, erosional, tectonic, and magmatic-related features that postdate the material. Noachian rocks are interpreted to consist mainly of impact breccias, volcanic materials (shields and flows), and eolian and fluvial sediments that mantle and subdue parts of the heavily cratered terrain. Densely cratered terrain is located to the south and southwest of the system of large valleys. Heavily and moderately cratered terrain also occurs to the north-

northeast of Valles Marineris, on the opposite side of the magmatic complex of Tharsis

Montes, several thousands of kilometers from the NSVs. Several of the major Noachian

rock units, which form the martian highlands, are observed in the region of interest

(Figure 2) including hilly (unit Nplh), cratered (unit Nph), ridged (unit Npk), and

subdued cratered (unit Npb) materials [e.g., Scott and Tanaka, 1986].

The hilly material at the base of the plateau sequence forms heavily cratered

terrain of high-standing plateaus, irregular topographic highs, prominent ridges, and

highly degraded, ancient crater rims mainly to the south of the NSVs and to the northeast

of Valles Marineris in the Xanthe Terra region. Locally, tectonic and impact processes

[Watters, 1993; Schultz and Tanaka, 1994] may have contributed to the prominent relief

of the hilly xinit The cratered material, vsrhich is the most extensive unit in the western

equatorial region [Scott and Tanaka, 1986; Tanaka, 1986], forms most of the basal rocks

to the south of the NSVs and partly embays the hilly material. The cratered material

records a high density of si5)erposed and partly buried and degraded impact craters of all sizes formed during the middle Noachian. Because of continued bombardment and

possible early, widespread volcanism {Saunders, 1979; Greeley and Spudis, 1981), the

cratered unit probably consists mostly of impact breccias and volcanic materials. The

ridged material outcrops to the south of the NSVs [e.g., Scott and Tanaka, 1986] and is

generally marked by rough, prominent, sublinear to irregular ridges. Generally, the

Noachian ridges are less continuous, less evenly spaced, and have more relief than

Hesperian ridges, but they follow similar regional trends, forming an arcmte pattern around the southern part of the Tharsis rise [e.g., Scott and Tamka, 1986; Schultz and

Tanaka, 1994].

Towards the end of the Late Noachian, many intercrater and intracrater areas of older materials were largely resurfaced by a thin mantle mapped as the subdued cratered material (unit Npl2) of the plateau sequence of mainly volcanic, eolian, and fluvial origin

[e.g., Scott and Tanaka, 1986]. The material embays most craters >10 km across and maiiily crops out far to tiie south of the NSVs and to the northeast of Valles Marineris in the Xanthe Terra region. It is marked in a few places by small ridges, channels, and possible flow fronts. Also during this time, centers of tectonic activity, many of which are interpreted to be the sites of magmatic-driven domal uplifts and associated volcanic eraptions, are documented near Arsia Mons (pre-dating the Tharsis Montes volcanoes),

Syria Planum, and the central part of Valles Marineris [Anderson et al, 1997, 1998,

1999, 2000; Dohm et al, 1998; Dohm and Tanaka, 1999) (Figures 1-4). Deep-seated magmatic bodies have been previously inferred for the western equatorial region where fault swarms have been deflected around their central cores such as at the central part of

Valles Marineris [Scott and Dohm, 1990]. The uplift of the Thaumasia plateau uplift may have also occurred during this time [Dohm and Tanaka, 1999] (Figures 1-4).

Comprehensive geologic investigations of the northwest part of the Thaumasia region indicate that magmatic/tectonic activity occurred at central Valles Marineris as early as the Late Noachian and diminished substantially during the Early Hesperian, prior to the emplacement of the younger ridged materials (unit Hr) [Dohm et al., 1998; Dohm and Tanaka, 1999]; such activity may have accompanied substantial volcanism. Similar 100 to the central part of Valles Marineris, Syria Planum is also a site of long-lived magmatic/tectonic activity, which includes domal uplift and associated radial and concentric faiilting during the Late Noachian-Early Hesperian, but at a much larger scale than is observed at the central part of Valles Marineris [Tanaka and Davis, 1988;

Andersonetal, \99%,\999,2Wi0-,DohmandTanaka, 1999).

Magmatic-related activity such as doming underlying Arsia Mons and located at

Syria Planum and central Valles Marineris during the Late Noachian/Early Hesperian may be genetically associated with the early development of the circum-Chryse outflow chaimel systems [e.g., Dohm et al, 1998; McKenzie and Nimmo, 1999). Importantly, such activity, which includes volatile and magma interactions, especially underlying

Arsia and at Syria, may have also contributed to the formation of the NSVs that occur to the south of Amazonis Planitia (Figures 1-4).

Hesperian System, The Hesperian System records extensive volcanism including the emplacement of wrinkle ridged plains materials (imit Hr) [e.g., Scott and Tanaka,

1986]. Wrinkle ridge materials dominate the Lunae Planum region located to the north of

Valles Marineris (Figure 2B) and outcrop, in places, south of the NSVs. In addition, widespread resurfacing of Early Hesperian and older materials, which includes the deposition of coUuvial and fluvial deposits and the dissection of rock materials by outflow activity north and northeast of Valles Marineris, is observed during the Early

Hesperian [e.g., Rotto and Tanaka, 1995; Nelson and Greeley, 1999]. Incipient Alba

Patera {Scott and Tanaka, 1986] and other centers of tectonic activity located in the Syria,

Pavonis (pre-Tharsis Montes), and Ulysses Fossae regions \Anderson et al, 1997,1998, 101

1999, 2000; Dohm and Tanaka, 1999), which are interpreted to be magmatic-related

domal uplifts, are also observed in the geologic record during this time; such activity is

indicative of growth of the magmatic complex of Tharsis (Figures 2 and 3).

Also during the Early Hesperian Period, outflow charmel development is recorded

at Mangala Valles located to the southwest of the NSVs [Craddock and Greeley, 1994;

Zimbelman et al, 1994] and at the circum-Chryse outflow chaimel systems located to the

north-northeast of Valles Marineris [e.g., Rotto and Tanaka, 1986; Nelson and Greeley,

1999], on the opposite side of the magmatic complex of Tharsis Montes several

thousands of kilometers fiom the NSVs; although significant outflow channel formation

is also recorded during the Late Hesperian and into the Early Amazonian such as at Kasei

{Chapman et al., 1991; Scott, 1993; Chapman and Tanaka, 1996] and Mangala

\Chapman and Tanaka, 1993; Craddock and Greeley, 1994; Zimbelman et al, 1994]

Valles.

Significant magmatic activity is also recorded during the Late He^rian and

Early Amazonian including the development of Olympus Mons and the Tharsis Montes

shield volcanoes [Scott and Tanaka, 1986; Morris et al, 1991; Morris and Tanaka, 1994;

Scott and Zimbelman, 1995; Scott et al, 1998] and the emplacement of voluminous sheet

lavas centered at the large shield volcanoes and at Syria Planum [Scott and Tanaka, 1986;

Dohm and Tanaka, 1999], which include members 1-2 of the Olympus Mons Formation,

members 1-4 of the Tharsis Montes Formation, and members 1 and 2 of the Syria Planum

Formation. Lava flows centered at Arsia Mons, for example, extend at least 1300 km to 102 the northwest {Zimbelman et al, 2000] partly embaying the gigantic northwest-trending promontories and partly infilling the NSVs (Figures 1 and 2).

Amazonian System, The continued development of Tharsis Montes and Olympus

Mons is recorded dviring the Amazonian Period including the emplacement of associated lavas and aureole deposits [Scott and Tanaka, 1986; Morris et al, 1991; Morris and

Tanaka, 1994; Scott and Zimbelman, 1995; Scott et al., 1998]. In addition, late-stage outflow chaimel development is observed at Mangala Valles [Chapman and Tanaka,

1993; Craddock and Greeley, 1994; Zimbelman et al, 1994], Marte Valles, which is interpreted to represent a spillway between the Elysium and Amazonis basins [Scott et al, 1995\, and the circxmi-Chryse outflow channel systems [Scott and Tanaka, 1986;

Chapman et al, 1991; Scott, 1993; Rotto and Tanaka, 1995; Chapman and Tanaka,

1996; Nelson and Greeley, 1999). Medusae Formation materials, which partly blanket the NSVs, form a broad but discontinuous band of wind-etched materials that trends east- west along the highland-lowland boundary [e.g., Scott and Tanaka, 1986; Greeley and

Guest, 1987; Scott and Chapman, 1991c]. On the basis of its morphologic characteristics, the Medusae Fossae Formation has been interpreted to consist of ash-flow tuffs [Malin, 1979; Scott and Tanaka, 1982, 1986; Zimbelman et al., 1997], ancient polar deposits [Schultz and Lutz, 1988], pyroclastic and eolian materials [Greeley and Guest,

1987], or a host of additional (but less likely) origins [Zimbelman et al, 1997].

Significant to the stiatigraphic history of the NSVs region, vrfiich has been clouded by late-stage volcanic and other activity, Sakimoto et al. [1999] have completed MOC and

MOLA investigations of previously mapped outcrops of Medusae Fossae Formation; they 103 indicate that the outcrops may be comprised of eolian or volcanic deposits that were emplaced on top of existing topography (e.g., older competent materials) and subsequently eroded. Similarly, we interpret the NSVs to be carved Noachian and possibly Early Hesperian materials draped by Late Hesperian and younger materials

(Figures 1-2). In addition, relatively small erosional terraces (Figure 1), located at the southwest and western parts of the NSVs, may indicate more recent, local fluvial activity, which is consistent with the recent identification of several small fluvial channels (Figure

5) that intermingle with the Medusae Fossae Formation materials in the region of the

NSVs [Zimbelman et al, 2000]. Such fluvial activity may be associated with late-stage development of Mangala Valles [Chapman and Tanaka, 1993; Craddock and Greeley,

1994; Zimbelman et al, 1994].

STRATIGRAPHIC SUMMARY OF THE NSVs

Noachian materials were faulted and carved by the Late Noachian-Early

Hesperian, forming a system of gigantic northwest-trending promontories and valleys that are associated stratigraphically with pre-Tharsis Montes magmatic-driven doming events of the Arsia, Syria Planum, and central Valles regions. The geometric shapes and geomorphic character of the gigantic, northwest-trending promontories, which include ghost craters, in comparison to other much smaUer outcrops of Medusae Fossae located to the west, indicate that the features are mostly comprised of promontory-forming, older materials. Late Hesperian and younger materials, which may include lavas, air fall deposits from Tharsis Montes, possible ash flow deposits, and eolian materials, appear to 104 blanket and embay the gigantic promontories and partly infill the valleys, subduing the relict landforms, especially when viewed fix>m Mariner and Viking images. Terraces located to the south of the system of valleys and, in places, along the valley floors may have resulted from local, late-stage fluvial activity. In addition, eolian, glacial, and mass- wasting processes may have modified the existing valley morphology.

DISCUSSION

In addition to the previously identified outflow channels, observations permitted by MOLA data have revealed the NSVs located along the highland-lowland dichotomy to the northwest of Arsia Mons and located south of Amazonis Planitia, a site of a postulated ocean [e.g, Parker et al, 1993] and (or) a paleolake [Scott et al., 1995], in the western hemisphere of Mars (Figures 1-4). Veneers of lava flows and volcanic air fall deposits, and possible ash flow deposits have partly obscured the NSVs. The veneers suppress evidence (e.g., relict features) making it difficult to rule out modes of valley origin. In addition to volcanic modification, local eolian, fluvial, and mass movement processes further obscure clues that would otherwise aid in better imderstanding the origin of the valleys.

Below, we present several hypotheses that attempt to explain the origin of the

NSVs. Following carefiil consideration of the stratigraphic, geomorphic, paleotectonic, topographic (using MOLA data), and geophysical data, we present these hypotheses, which include catastrophic flooding, tectonism, mass wastiug, glaciation, and wind erosion. It is important to keep in mind that combinations of any of the activities listed 105 below, occurring concurrently or at different times, may have contributed to valley formation.

Hypothesis 1: Catastrophic Flooding. The premise that the NSVs located near the western slope of Tharsis resulted from one or more catastrophic floods assumes that magmatic-driven processes triggered sudden releases of large volxmies of water into

Amazonis Planitia (forming a watershed to the northwest of the Arsia/Syria dome complex) (Figures 1-4), similar to the driving mechanism often envisioned for the formation of the smaller circum-Chryse outflow channel systems that debouch into

Chryse Planitia. Spatial and temporal associations of the NSVs with magmatic-driven.

Late Noachian to Early Hesperian doming events located near Arsia (pre-Tharsis Montes) and Syria Planum, may corroborate such activity. During the same time, tremendous volumes of water were issued into Chryse Planitia from source areas near Valles

Marineris, a region of recorded magmatic and tectonic activity. Similar to circum-Chryse outflow channel systems, the implications of uncovering such a significant flood record of the ancient martian past is of great significance in that such activity supports northern ocean(s) and (or) large paleolakes in the northern plains.

While observations of the NSVs do not individually confirm Hypothesis 1, collectively, they corroborate tremendous flooding. Some of the more significant observations include:

• the tremendous size of the ^stem of valleys and promontories,

• the unique shape of the gigantic promontories with respect to smaller

topogr^hic highs located to the west of the NSVs, including tear-dropped 106 shaped ends and tremendously long, relatively straight margins; the shapes indicate that the giant outcrops mostly comprise competent materials similar to the flood carved mesas north and east of Valles Marineris (Figure 1), terraces, in places, mark the margins of the gigantic promontories, a distinctively smooth surface located directly north of the NSVs in Amazonis

Planitia, along the western margin of Olympus and associated lava flows (in contrast to the knobby terrain to the west) that closely resembles sub-aqueoxis sedimentation in abyssal plains and sedimentary basins (Aharonson et al.,

1998) or flood-scoured desert terrains on Earth, the system of valleys that generally correspond spatially to gravity lows and occur within a large topographic depression (Figure 2), similar to the eastem

Chiyse Planitia outflow channels [Phillips et al., 2000]; the gravity lows may represent low-density materials that partly infill the valleys and large topographic depression, the Late Noachian-Early Hesperian age of the NSVs that corresponds stratigraphically with the early outflow development of the circum-Chryse outflow channel systems and with magmatic-driven, pre-Tharsis doming events, which occurred upgradient and to the southeast of the NSVs near

Arsia Mons and at Syria Planum (Figures 1-4), and the trends of the gigantic outcrops, especially the southeast parts in conjunction with linears such as scarps/terraces located to the northeast of the 107

promontories (Figures 1 and 3), which point towards the potential source

region, Arsia Mons (pre-Tharsis Montes) and Syria Planiun.

Unlike the development of the circum-Chryse outflow channel systems, which may have occurred from the Late Noachian to Early Amazonian, stratigraphic evidence suggests that potential northwestern slope flooding was limited to the Late Noachian-

Early He^rian, possibly cut off from a massive water supply by a drainage divide formed by a fiilly developed Syria/Arsia dome complex and (or) late-stage growth of the

Tharsis Montes shield volcanoes; although minor, local fluvial activity occurred later in the southern and western parts of the NSVs. As with the occurrence of mass wasting (see

Hypothesis 3), it is important to realize that catastrophic flood events are not necessarily limited to streamflow. There is little doubt that the NSVs signify multiple flood events of various m^nitudes prior to the Late Hesperian. This raises the possibility that phases of hyperconcentrated-flow and debris-flow activity may have been associated with the flooding in this area. Such variations in flow behavior would create a unique assemblage of deposits [Webb et. al., 1989] that may have at least partly filled these canyons between flood events.

Hypothesis 2: Tedonism. Tectonic structural control may have played a significant role in channeling catastrophic flood waters fix)m the Arsia SW/Syria dome complex region (Figure 4) to the Amazonis Planitia region, resulting in significant enhancement of the original valley geometry of the NSVs. For that reason, it is important not to understate the importance of this hypothesis in relation to catastrophic flooding.

Large-scale tectonics fabrics, for example, exist within and adjacent to the NSVs. These 108 include; (1) ridged material, which crops out to the south of the NSVs and is comprised of sublinear to irregular ridges that exhibit trends common to the NSVs, (2) faults that cut

Noachian surfaces to the southwest of the special site of interest, and (3) large-scale tectonic fabrics adjacent and within the NSVs, such as Gordii Dorsum [e.g., Scott and

Tanaka, 1986]. Although tectonic features may have routed the erosional agent, tectonic activity by itself probably did not result in the gigantic size of the NSVs.

Hypothesis 3: Mass wasting. The formation of outflow channels emptying into the Chryse Planitia on Mars has been attributed to debris flow activity [Tanaka, 1999].

Mass movements are common in the Valles Marineris canyon system and may be responsible for the development of the aureole deposits on the flanks of Olympus Mons

[Hodges and Moore, 1994]. Therefore, the possibility that mass wasting may have created the NSVs must be thoroughly examined.

Debris flows on Earth are capable of moving hv^e quantities of material great distances (more than 800 km in the subaqueous Storegga slide) [Bugge et. al, 1988].

Most debris flows move in existing topographic depressions, however, and may occur as a brief phase accompanied by subsequent streamflow or hyperconcentrated flow [Mellis et. al, 1997]. Typically mass movements will create a scarp or headwall in their source region where the slide material is removed and a lobate zone of deposition where the material stops moving. Erosion is greatest at the source area where debris-flow activity has been observed; in some cases, such activity may remove all of the existing unconsolidated material in the source area, leaving exposed bedrock [Johnson and

Rodine, 1984]. Landslide deposits such as debris-flow levees, slump-blocks, or a pile of 109

rubble produced by a rock topple are the most typical landfonns produced during these

mass movements.

No direct evidence of deposits created by mass movement can be found in or

around the NSVs. In addition, no scarps or headwalls are visible near Arsia Mons where

either mass movements or floodwaters would have originated. Higher resolution images

of the floors of the NSVs and Amazonis Planitia should be examined for deposits typical

of mass movements, especially debris-flow deposits which would be the most likely sub-

aerial mass movements to travel great distances from their source. Sub-aqueous

landslides have the longest recorded runouts of any mass movements on Earth [Bugge et

al, 1988]. If the area north of the NSVs were filled with water at any time in martian

geologic past, sub-aqueous landslides covdd have moved trough the area, moving large

amounts of material out of the channels and into Amazonis Planitia. Such sub-aqueous

slides on Earth have moved huge, intact blocks of material great distances from the

slide's source region \Bugge et al, 1988]. Similar activity on Mars could deposit large

blocks at great distances out into the smooth plains north of the proposed channels.

It is not likely that mass movement activity alone created the NSVs. Depositional

or erosional features typically associated with mass movements are not visible in the

region of special interest In addition, mass movements usually concentrate their erosive activity in their source region and thus would not be likely to create channels a significant distance downslope from their source region. Mass movements are expected as part of

the streamflow that may have created these channels. Post-flood slumping could have 110 occurred on the sides of the valleys and some debris flow activity may have occurred in the valleys themselves as a phase in an intermittent streamflow regime.

Hypothesis 4: Glaciation. Ice streams have been proposed to have carved the circum-Chryse outflow channel systems [e.g., Lucchitta and Anderson, 1980]. In order for glaciers to have formed the NSVs, the climate had to differ significantly firom the present The "MEGAOUTFLO" hypothesis first demonstrated by Baker et al. [1991] genetically links magmatic-triggered activity with relatively short-lived climatic perturbations firom a common cold and dry Mars (e.g., short-term hydrological cycles), which includes enhanced erosion by precipitation, landslides/mass wasting (both land and submarine), glacial activity, and the presence of large standing bodies of water such as the Amazonian-aged Oceanus Borealis. A synthesis of the stratigraphic, erosional, and paleotectonic records indicate potential pulses of magmatic activity may have triggered climatic perturbations such as diaing the Late Noachian-Early Hesperian [Dohm et al,

2000] when the NSVs were carved. In addition, in order for glaciers to develop and move down the northwest flank of the pre-Tharsis Montes, magmatic-driven Arsia uplift, a plentifiil supply of water must have been present. This water could be provided through an Earth-like hydrologic cycle, which may have existed episodically on Mars [Baker et. al, 1991; Baker et al, 2000], or subsurface water that fijoze iQ)on contact with the surface, creating an ice stream [Lucchitta et. al, 1981, Lucchitta, 1982].

Glacier formation by accumulation of snow through a hydrologic cycle assumes that precipitation (most likely orographic) would have resulted fixDm prevailing winds rising over the Arsia uplift Such precipitation woidd have to continue over a sufficient Ill period of time for the necessary thickness of ice to buildup for glacial flow to initiate.

Interpretations of striations on the southwestern flank of Arsia Mons as glacial moraines

{Hodges and Moore, 1994] and fan-shaped deposits on the northwest flanks of the

Tharsis Montes volcanoes, which consist of facies that are interpreted to be the result of glacial and volcanic^^ce interactions [Scott and Zimbelman, 1995; Scott et ah, 1998], indicate that variations in climate from the present one have occurred in the recent geologic past. These features, however, are dwarfed by the system of gigantic valleys that formed much earlier in martian time, during the Late Noachian-Early He^rian.

The major drawback to the glacial-origin hypothesis is the fact that no characteristic features resulting from glacial erosion or deposition are seen in the region encompassing the NSVs. Depositional features such as eskers, drumlins, or moraines are absent as are erosional features such as cirques at the head of the valleys or aretes along the ridgelines between the valleys. Particularly, there are no moraines or eskers down gradient from the NSVs. An explanation for this is that all evidence of glaciation, except for the gigantic promontories, have been buried and (or) obscured by materials (including volcanic, eolian, colluvial, possible marine, and possible lacustrine) or destroyed by wind and water erosion.

Hypothesis 5: Wind Erosion. During an absence of an active hydrologic cycle for much of the history of Ihe planet {Baker et ah, 2000], wind erosion played an important role in the modification of the surface of Mars {Greeley et. al., 1999]. Thus, it is important to consider the possibility that the NSVs are the result of wind erosioiL On 112

Earth, winds are eflScient at sculpting existing surfaces, but not sufficient to carve bedrock canyons of the magnitude observed for the NSVs.

Wind erosion would be most effective in eroding unconsolidated surficial deposits, so the competence of the material in the study area is a major factor to take into account in the assessment of the efiScacy of wind erosion. The unique shape of the gigantic promontories with respect to smaller topographic highs mapped as Medusae

Fossae Fonnation materials located to the west of the region of interest, however, indicate that the gigantic landfoims comprise competent materials similar to the flood carved mesas north and east of Valles Marineris (Figure 1). Evidence of recent wind activity exists in and around the special site of interest, especially yardangs and wind streaks that mark the Medusae Fossae Formation surfaces [e.g., Scott and Chapman,

1991c]; the wind streaks provide evidence of winds originating from several different directions in recent times. Wind erosion is an unlikely sole candidate for the fonnation of the NSVs for the following reasons:

• wind erosion on Earth lacks the power needed to be the sole erosive force for

carving gigantic bedrock valleys,

• wind-formed features of a much smaller scale than the gigantic promontories

indicate a variety of wind directions for the special region of interest whereas

a single vmd direction would have to be maintained a long period of time for

efficient wind erosion to occur, and

• a well-established, strong wind pattern in this area capable of large-scale

erosion would have produced other features of similar size in the area at the 113

same time. No such wind-related features of gigantic proportions can be

found elsewhere on Mars.

PALEOHYDROLOGY

After careful consideration of the previously mentioned processes, structurally controlled flooding appears to best explain the formation of the NSVs, although the other processes probably further modified the NSVs. Here we attempt to determine the ancient hydrologic conditions based on our knowledge of the geometry of the valleys, as they exist today. This task, however, is daunting due to subsequent volcanic, eolian, fluvial, and gravity-driven modification of the NSVs. Extensive masking of the walls of the

NSVs by the Medusae Fossae Formation means that the earlier fluvial erosion of whole channel cross sections must be inferred in order to perform the calculations that follow.

The paleohydrology of the NSVs was estimated fi-om an analysis of the geometry of the valleys (e.g., several MOLA-based valley profiles were generated and analyzed).

There are limitations imposed on these calculations. The ancient flow depths, for example, can only be approximated because volcanics as weU as other materials of unknown thickness obscure the original valley geometries. Thus, we sought to determine the sensitivity of the discharge value by altering the constituent variables of the modified

Manning equation [Sellin, 1969; Komar, 1979]:

W=[(gm*A*5)/Cf]^ 114

where u is the average flow velocity, gn, is the Martian gravitational acceleration, h is the

hydraulic radius, s is the sine of the channel-bed slope, and Cf is a dimensionless drag

coefiBcient approximated by:

Cf=ge*(«'/h''^)

where ge is the gravitational acceleration of Earth and n is the Manning roughness

coefiicienL Using the method of Komar (1979), discharge calculations for the NSVs

were made (Table 1).

The greatest source of error in the Manning equation is the value of the Manning

roughness coefficient After careful consideration of hydrologic and geologic factors, a

value of 0.025 was selected for the Manning roughness coefficient. It is interesting to

note that this value, determined independently, s^ees with the value used by Robinson

and Tanaka [1990] in the calculation of flood discharge for Kasei Valles, an outflow- channel system that is generally agreed to be the largest currently knoAvn on Mars. 115

Table 1. A comparative analysis between some of the previously defined martian outflow valley systems and the newly defined northwestern slope valleys (NSVs).

Reference Channel D(m) Wflan> S Vfm/s) O fm'/s)

Komar [1979] Mangala 100 14 0.003 15 2x10^

Baker [\9%2\ Maja 100 80 0.02 38 3x10®

Komatsu and Ares 500-1000 25 0.02 - 0.0001 25 -150 5x10® Baker [1997]

Robinson and Kasei 400 -1300 80 0.009 30-75 1-2x10^ Tanaka [1990]

Dohm et al NSVs 1200-2400 100 - 700 0.004 - 0.005 30-40 -10^-10^° [in press]

A series of calculations were made to determine what effect certain hydrologic variables would have on the discharge values. For example, we performed numeroxis alterations to the valley geometries including: (1) lowering the valley floors by intervals of 100 meters until reaching a depth of one kilometer, in order to approximate valley profiles prior to volcanic infilling of NSV, (2) decreasing the valley widths by 20 and 40 percent, to account for the possible removal of canyon wall material by wave-cut action, erosion, and (or) mass movements, (3) modifying the valley slopes, in order to accurately gage the sensitivity of the calculations to the shallowing of the regional slope due to infilling volcanic materials, and (4) changing the water depth fix>m a bare TninimTim amount needed to cover the channel floor (roughly 10% maximum capacity), up to 100 percent of the NSV's maximum capacity. Although the valley geometries were modified 116 in almost every conceivable way possible, due to the immense size of this valley system, the discharge values remained consistently large.

More specifically, in the calculation of discharge values, the valley profiles were corrected to reflect the infilliDg of volcanic and other materials since the Late Hesperian

(e.g.. Figure 2C). Progressively away from Arsia Mons, the valley floors were lowered

500 m, 250 m, and 0 m, respectively. The modified geometry of the channel system and a Manning roughness coefficient value of 0.025 yielded a maximum discharge of 2 x I0'° mVs. The discharge value for an assumed flow depth is summarized in the following graph:

2.5E+10

2.0E+10

^ 1.5E+10

1.0E+10

0 10 20 30 40 50 60 70 80 90 100 Percent Filled

The most interesting result of these calculations is that no matter what modifications were made to the observed valley geometries, whether it be the narrowing 117 of the valleys, the lowering of the valley floors, altering the regional slope, or assvaning the NSVs was never filled by more than 10 percent of its maximiun capacity, the resultant discharge values remain large. All other calculated discharge values for the circum-Chryse outflow channel systems of the northeastern watershed of the Thaisis region combined fail to equal many of the maximum discharge estimates for the NSVs.

The NSVs potentially represent catastrophic flooding from an ancient (Late

Noachian-Early Hesperian) watershed located to the northwest of the Arsia/Syria dome complex, perhaps related to the early development of the circum-Chryse outflow channel systems, as well as a potential source of water for a northern plains ocean. Two shorelines have been proposed for the northern plains; an inner younger one that is close to an equipotential line, and an older one that deviates from an equipotential line [Parker et al., 1993; Head et al., 1999]. The inner younger ocean is estimated to have had a volume of about 1.4x10^ km^ and the larger older one a volume of about 9.6xl0' km^

[Head et al., 1999]. The ancient topography, however, most likely varied from the present one. Thus, the estimated calculated volume of the putative larger ocean may be considerably different. It is interesting to note that for the maximum discharge rate of

2x10^° mVs, assuming the flow rate was sustained, the fill time would have been about

8.1 days and about 8 weeks for the smaller and larger oceans, respectively. If the Chiyse outflows wCTe occurring at the same time at rates of about lO' m^/s to 10'° m^/s then the fill rates would be considerably shorter, fix>m about 7.7 to 5.4 days for the smaller ocean.

The discharge rates, however, were most likely not sustained at these peak values and a greater time period would be required to fill these hypothesized oceans. In addition. 118

other hydrogeologic activities associated with catastrophic flooding and related short­

lived (—10'* to 10^ year) episodes of quasi-stable climatic conditions (Baker et al., 1991;

Baker et al., 2000) may have also contributed water to the hypothesized oceans by mechanisms such as q)ring-fed activity along areas of the highland-lowland boundary.

Another potential contributor to the northern plains ocean is a large quantity of groimd

ice (e.g., stagnant ice sheets) already in place in the northern plains during this time. This large quantity of ice would be the likely consequence of an earlier warm and wet phase of

Mars, possibly induced by catastrophic flooding (Baker et al., 1991; Baker et al., 2000).

As new floodwaters washed over the northern plains, the additional heat would melt the

upper layers of ice and the gradients created would allow the melt water to cycle into the hydrologic system.

IMPLICATIONS

Although several processes were most likely involved with the present-day valley geometries of the NSVs, catastrophic flooding prior to Late Hesperian and younger volcanism (including other depositional and erosional modification) contributed significantly to the formation of the NSVs. The implications for flood-carved NSVs are enormous; The NSVs potentially represent: (1) previously imdocimiented maitian catastrophic floods, (2) a watershed to the northwest, peifaaps related to the early development of the circum-Chiyse ^stems of outflow channels, and (3) a potential source of water for a northern plains ocean. The discharge rates have been estimated between about 10* and 10'° mVs. Even if maintained for only a short time, these rates would have significantly contributed to the formation of a northern plains ocean. 119

REFERENCES

Aharonson, O., M. T. Zuber, and G. A. Neumann, Mars: Northern hemisphere slopes and slope distribxrtions, Geophys. Res. Lett., 25,4413-4416,1998.

Anderson, R. C., M. P. Golombek, B. J. Franklin, K. L. Tanaka, J. M. Dohm, J. H. Lias, and J. Higdon, A new detailed structural history of the Tharsis region of Mars, (abstract). Lunar Planet ScL Conf., XXVIH, 39-40,1997.

Anderson, R. C., M. P. Golombek, B. J. Franklin, K. L. Tanaka, J. M. Dohm, J. H. Lias, and B. Peer, Centers of tectonic activity through time for the western hemisphere of Mars, Lunar Planet. ScL Conf. [CD-ROM], XXDC, abstract 1881,1998.

Anderson, R. C., A. F. C. Halderman, J. M. Dohm, M. P. Golombek, B. J. Franklin, and J. H. Lias, Significant centers of tectonic activity as identified by wrinkle ridges for the western hemisphere of Mars, Lunar Planet Sci. Conf. [CD-ROM], XXX, abstract 1972,1999.

Anderson, R. C., and J. M. Dohm, Magmatic-tectonic evolution of Tharsis, Lunar Planet Sci. Conf [CD-ROM], XXXI, abstract 1607,2000.

Baker, V.R,, The Spokane Flood controversy and the Martian oiitflow chaimels. Science, 202,1249-1256,1978.

Baker, V. R., The channels of Mars, Austin, University of Texas Press, 198 pp, 1982.

Baker, V. R., and D. J. Milton, Erosion by catastrophic floods on Mars and Earth, Icarus, 23,27-41,1974.

Baker, V. R., R. G. Strom, V. C. Gulick, J. S. Kargel, G. Komatsu, and V. S. Kale, Ancient oceans, ice sheets and the hydrological cycle on Mars, Nature, 352,589-594, 1991. 120

Baker, V. R., R. G. Strom, J. M. Dohm, V. C. Gulick, J. S. Kargel, G. Komatsu, G. 0. Ori, and J. W. Jr. Rice, Mars' Oceanus Borealis, ancient glaciers, and the MEGAOUTFLO hypothesis. Lunar Planet. Sci. Conf. [CD-ROM], XXXI, abstract 1863,2000.

Bretz, J H., The Lake Missoula floods and the Channeled Scabland, J. Geology, 77, 505-543,1969.

Bugge, T., R. H. Belderson, and N. H. Kenyon, The Storegga slide, Phil. Trans. R. Sac. Land., 325,357-388,1988.

Carr, M. H., Formation of Martian flood features by release of water from confined aquifer,^ Geophys. Res., 84,2995-3007,1979.

Carr, M. H., R. Greeley, K. R. Blasius, J. E. Guest, and J. B. Murray, Some martian volcanic features are viewed from Viking Qrbiters, J. Geophys. Res., 82, 3985-4015, 1977.

Chapman, M.G., Evidence, age, and thickness of a frozen paleolake in Utopia Planitia, Mars, Icarus, 109,393-406,1994.

Chapman, M. G., H. Masursky, and D. H. Scott, Geologic map of science study area 2, north Kasei Valles, Mars (MTM 25072 quadrangle), USGS Misc. Inv. Ser. Map I- 2107 (1:500,000), 1991.

Chapman, M. G., and K. L. Tanaka, Geologic map of the MTM -05152 and -10152 quadrangles, Mangala Valles region of Mars, USGS Misc. Inv. Ser. Map 1-2294 (1:500,000), 1993.

Chapman, M. G., and K. L. Tanaka, Geologic maps of the MIM 25062 quadrangle (digital compilation) and the MTM 25067 quadrangle (manual compilation), Kasei Valles region of Mars, USGS Misc. Inv. Ser. Map 1-2398 (1:500,000), 1996. 121

Craddock, R. A., and R. Greeley, Geologic map of the MTM -20147 quadrangle, MangalaValles region of Mars, USGSMisc. Inv. Ser. Mcq) 1-2310 (1:500,000), 1994.

Crown, D. A., K. H. Price, and R. Greeley, Geologic evolution of the east rim of the Hellas basin. Mars. Icarus, 100,1-25,1992.

Crown, D. A., and R. Greeley, Volcanic geology of Hardriaca Patera and the eastern Hellas region of Mars, J. GeopJiys. Res., 98,3431-3451,1993.

DeHon, R. A,, Geologic map of the Pompeii quadrangle (MTM 20057), Maja Valles region of Mars, USGS Misc. Inv. Ser. Map 1-2203 (1:500,000), 1992.

Dohm, J. M., R. C. Anderson, and K. L. Tanaka, Digital structural mapping of Mars, Astronomy & Geophysics, 39,3.20-3.22, 1998.

Dohm, J, M., and K. L. Tanaka, Geology of the Thaumasia region. Mars: plateau development, valley origins, and magmatic evolution. Planetary and Space Science, 47,411-431,1999.

Dohm, J. M., R. C. Anderson, V. R. Baker, R. G. Strom, G. Komatsu, and T. M. Hare, Pulses of magmatic activity through time: potential triggers for climatic variations on Mars, Lunar Planet. Sci. Conf. [CD-ROM], XXXI, abstract 1632,2000.

Dohm, J. M., R. C. Anderson, V. R. Baker, J. C. Ferris, T. M. Hare, R. G. Strom, L. P. Rudd, J. W, Jr. Rice, and D. H. Scott, System of gigantic valleys northwest of Tharsis Montes, Mars: Latent catastrophic flooding, northwest watershed, and implications for northern plains ocean, Geophys. Res. Lett., in press.

Greeley, R., and P. D. Spudis, Volcanism on Mars, Rev. Geopf^s. Space Phys., 19, 13- 41,1981.

Greeley, R, and J. E, Guest, Geolo^c map of the eastem equatorial region of Mars, USGSMisc. Inv. Ser. MapI-1802B (1:15,000,000), 1987. 122

Greeley, R., M. Kraft, R. Sullivan, G. Wilson, N. Bridges, K. Herkenhofif, R. 0. Kxizmin, M. Malin, and W. Ward, Aeolian features and processes at the Mars Pathfinder landing site, J. Geophys. Res., 104,8573-8584,1999.

Gregg, T. K. P., and P. H. Schultz, Shallow Martian "sills": Intrusions or extrusions?. Lunar Planet. Set Conf., XXVUI, 463-464,1997.

Head HI, J. W., H. Hiesinger, M. A. Ivanov, M. A. Kreslavsky, S. Pratt, and B. J. Thompson, Possible ancient oceans on Mars: evidence from Mars Orbiter Laser Altimeter data. Science, 286,2134-2137,1999.

Hiller, K., Geologic map of the Amenthes quadrangle of Mars, USGS Misc. Im. Ser. Map I-lllO (1:5,000.000), 1979.

Hodges, C. A, and H. J. Moore, Atlas of Volcanic Landforms on Mars, U.S. Geol. Surv. Prof. Pap., 1534, 194 pp., 1994.

Johnson, A. M. and J. R Rodine, Debris flows, in D. Brunsden, and D. B. Prior, Slope Instability, pp. 257-361, New York, John Wiley and Sons, 1984.

Jons, H. P., Arcuate groxmd undulations, gelifluxion-like features and "front tori" in the northern lowlands on Mars, what do they indicate?, (abstract). Lunar. Planet. Sci. Conf., XVn, 404-405,1986.

Kargel, J. S., V. R. Baker, J. E. Beget, J. F. Lockwood, T. L. Pewe, J. S. Shaw, and R. G. Strom, Evidence of ancient continental glaciation in the martian northern plains, J. Geopl^s. Res., 100,5351-5368,1995.

Kargel, J. S., K. L. Tanaka, V. R Baker, G. Komatsu, and D. R. MacAyeal, Formation and dissociation of clathrate hydrates on Mars: polar caps, northern plains, and highlands, itmarP/flwet Sci. Conf. [CD-ROM], XJOQ, abstract 1891,2000. 123

Komar, P. D., Comparisons of the hydraulics of water flows in Martian outflow channels with flows of similar scale on Earth, Icarus, 37, 156-181, 1979.

Komatsu, G. and V. R. Baker, Paleohydrology and flood geomorphology of Ares Vallis, J. Geophys. Res., 162,4151-4160,1997.

Komatsu, G., J. S. Kargel, V. R. Baker, R. G. Strom, G. G. Ori, C. Mosangini, and K. L. Tanaka, A chaotic terrain formation hypothesis: explosive outgas and outflow by dissociation of clathrate on Mars, Lunar Planet. Sci. Conf. [CD-ROM], XXXI, abstract 1434,2000.

Lucchitta, B. K., Ice sculpture in the Martian outflow channels, J. Geophys. Res., 87, 9951-9973,1982.

Lucchitta, B. K., Recent mafic volcanism on Mars, Science, 235,565-567,1987.

Lucchitta, B. K. and D. M- Anderson, Martian outflow channels sculptured by glaciers, (abstract), in Reports of Planetary geology program - 1980, NASA TM-81776, 271- 273,1980.

Lucchitta, B. K., D. M. Anderson, and H. Shoji, Did ice streams carve Martian outflow channels?. Nature, 290,759-763,1981.

Lucchitta, B. K., H. M. Ferguson, and C. Summers, Sedimentary deposits in the northern lowland plains, Mars, J. Geophys. Res., 91, E166-E174,1986.

Lucchitta, B. K-, A. S. McEwen, G. D. Clow, P. E. Geissler, R. B. Singer, R. A. Schultz, and S. W. Squyres, The canyon system on Mars, in Mars, pp. 453-492, Univ. of Ariz. Press, Tucson, 1992.

MacKinnon, D. J., and K. L. Tanaka, The impacted martian crust: structure, hydrology, and some geologic implications, J. Geophys. Res., 94,17,359-17370,1989. 124

Malin, M. C., Mars: evidence of indurated deposits of fine materials, (abstract), NASA Conf. Pub. 2072, 54,1979.

Mars Channel Working Group, Channels and valleys on Mars, Geol Soc. Amer. Bull, 94,1035-1054,1983.

Masursky, H., J. M. Boyce, A. J. Dial, G. G. Schaber, and M. E. Strobell, Formation of Martian channels, J. Geophys. Res., 82,4016-4038,1977.

Max, M. D., and S. M. Clifford, The identification of Martian outflow channels through the catastrophic decomposition of methane hydrate. Lunar Planet. Sci. Conf. [CD- ROM], XXXI, abstract 2094,2000.

McGill, G. E., Age and origin of large martian polygons, (abstract). Lunar. Planet. Sci. Conf., XVI, 534-535,1985.

McKenzie, D., and F. Nimmo, The generation of Martian floods by the melting of ground ice above dykes. Nature, 397,231-233,1999.

Melis, T. S., R. H. Webb, and P. G. Griffiths, Debris flows in Grand Canyon National Park: Peak discharges, flow transformations, and hydrographs, in Debris-flow Hazards Mitigation: Mechanics, Prediction and Assessment, ^ted by Chen-lung, New York, Amer. Soc. of Civil Eng., 817 pp., 1997.

Milton, D. J., Geologic map of the Limae Palus quadrangle of Mars: USGS Misc. Inv. Ser. Map 1-894 (1:5,000,000), 1974.

Morris, E. C., H. Masursky, D. J. Applebee, and M. E. Strobell, Geologic map of science study area 3, Olympus Rupes, Mars (special MTM 15132 quadrangle), USGS Misc. Inv. Ser. Map 1-2001,1991.

Morris, E. C., and K. L. Tanaka, Geologic map of the Olympus Mons region of Mars, USGS Misc. Inv. Ser. Map 1-2327,1994. 125

Nelson, D. M., and R. Greeley, Geology of Xanthe Terra outflow channels and the Mars Pathfinder landing site, J. Geophys. Res., 104,8653-8669,1999.

Nummedal, D. and D, B. Prior, Generation of Martian chaos and channels by debris flows, Icarus, 45,77-86, 1981.

Parker, T. J., D. M. Schneeberger, D. C. Pieri, and R. S. Saunders, Geomorphic evidence for ancient seas on Mars, in Symposium on Mars: Evolution of its Climate and Atmosphere, LPITech. Rept. 87-01,96-98, 1987.

Parker, T. J., D. S. Gorsline, R. S. Saunders, D. C. Pieri, and D. M. Schneeberger, Coastal geomorphology of the martian northern plains, J. Geophys, Res., 98, 11,061-11,078, 1993.

Phillips, R. J., M. T. Zuber, S. A. Hauck H, R. M. Williams, and K. B. Portle, Why is there a negative gravity ring around Tharsis on Mars?, Lunar Planet. Sci. Conf. [CD- ROM], XXXI, abstract 1303,2000.

Plescia, J.B., Recent flood lavas in the Elysium region of Mars, Icarus, 88, 465-490, 1990.

Plescia, J. B., and R. S. Saunders, Tectonic history of the Tharsis region. Mars, J. Geophys. Res., 87,9775-9791,1982.

Price, K. H., Geologic map of the Dao, Harmakhis, and Reull Valles region of Mars, USGSMisc. Inv. Ser. Map 1-2557 (1:1,000,000), 1998.

Rice, J. R., Jr., and R. A. DeHon, Geologic map of the Darvel Quadrangle (MTM 20052), MajaValles region of Mars, USGSMisc. Inv. Ser. Map 1-2432 (1:500,000), 1996.

Robinson, M. S., and K. L. Tanaka, Magnitude of a catastrophic flood event at Kasei Valles, Mars, Geology, 18,902-905,1990. 126

Rotto, S.L., and K. L. Tanaka, Geologic/geomorphologic map of the Chryse Planitia region of Mars, USGS Misc. Inv. Ser. Map 1-2441, 1995.

Sakimoto, S. E. H., H. V. Frey, J. B. Garvin, and J. H. Roark, Topography, roughness, layering, and slope properties of the Medusae Fossae Formation from Mars Orbiter Laser Altimeter (MOLA) and Mars Orbiter Camera (MOC) data, J. Geophys. Res., 104,24,141-22,154,1999.

Saunders, R. S., Geologic map of the Margaritifer Sinus quadrangle of Mars, USGS Misc. Inv. Ser. Map 1-1144 (1:5,000,000), 1979.

Schultz, P. H., and A. B. Lutz, Polar wandering on Mars, Icarus, 73,91-141, 1988.

Schultz, R. A., Geologic map of the western Ophir Planum region (MTM-10067) of Mars, USGS Misc. Inv. Ser. Map 1-2588 (1:500,000), 1998.

Schultz, R. A., and K. L. Tanaka, Lithospheric-scale buckling and thrust structures on Mars: The Coprates rise and south Tharsis ridge belt, J. Geophys. Res., 99, 8371- 8385,1994.

Scott, D. H., Mars, highlands-lowlands: Viking contribution to Mariner relative age studies, Icarus, 34,479-485, 1978.

Scott, D. H., Geologic map of the MTM 25057 and 25052 quadrangles, Kasei Valles region of Mars, USGS Misc. Inv. Ser. I-Map 2208 (1:500,000), 1993.

Scott, D. H., and K. L. Tanaka, Ignimbrites of the Amazonis Planitia region of Mars, J. Geophys. Res., 87,1179-1190,1982.

Scott, D. H., and K. L. Tanaka, Geologic map of the western equatorial region of Mars, USGS Misc. Inv. Ser. MapI-1802-A (1:15,000,000), 1986. 127

Scott, D. H., and J. M. Dohm, Tectonic setting of Martian volcanoes and deep-seated intrusives. In MEVTV Workshop "Evolution on Magma Bodies on Mars," 39-40, 1990.

Scott, D. H., M. G. Chapman, J. W. Jr. Rice, and J. M. Dohm, New evidence of lacustrine basins on Mars: Amazonis and Utopia Planitia, in Proc. Lunar. Planet. Set. 22, 53- 62,1991a.

Scott, D. H., J. W. Jr. Rice, and J. M. Dohm, Martian paleolakes and waterways; Exobiological implications, in Origins of Life and Evolution of the Biosphere 21,189- 198,1991b.

Scott, D. H., and M. G. Chapman, Geologic map of science study area 6, Memnonia region of Mars (MTM —10172), USGS Misc. Inv. Ser. Map 1-2084 (1:500,000), 1991c.

Scott, D. H., J. M. Dohm, and D. J. Applebee, Geologic map of science study area 8, Apollinaris Patera region of Mars, USGS Misc. Inv. Ser. Map 1-2351 (1:500,000), 1993.

Scott, D. H. and M. G. Chapman, Geologic and topographic maps of the Elysium Paleolake Basin, Mars, USGS Misc. Inv. Ser. Map 1-2397 (1:5.000,000), 1995.

Scott, D. H., J. M. Dohm, and J. W. Jr. Rice, Map of Mars showing channels and possible paleolake basins, USGS Misc. Inv. Ser. Map 1-2461 (1:30,000,000), 1995.

Scott, D.H. and J. R. Zimbelman, Geologic map of Arsia Mons volcano. Mars, USGS Misc. Inv. Ser. Map 1-2480 (1:1,000,000), 1995.

Scott, D. H., J. M. Dohm, and J. R. Zimbelman, Geologic map of Pavonis Mons volcano. Mars, USGS Misc. Inv. Ser. Mcp 1-2561 (1:1.000,000), 1998.

Sellin, R. H. J., Flow in channels, London, Macmillan, 149 pp., 1969. 128

Sleep, N. H., Martian plate tectonics, J. Geophys. Res., 99,5,639-5,655,1994.

Tanaka, K. L., The stratigraphy of Mars, Proc. Lunar Planet. Sci. Conf. 17th, in J. Geopf^s. Res.,pt. 1,91, E139-E158,1986.

Tanaka, K. L., Debris-flow origin for the Simud/Tiu deposit on Mars, J. Geophys. Res., 104,8637-8652,1999.

Tanaka, K. L., and P. A. Davis, Tectonic history of the Syria Planum province of Mars, Geophys. Res., 93,14,893-14,917,1988.

U. S. Geological Survey, Controlled photomosaic of the MTM 05142 quadrangle, Gordii Dorsum region of Mars, U. S. Geol. Siirv. Misc. Inv. Ser. Map 1-2425, scale 1:500,000,1995.

Walters, T. R., Compressional tectonism on Mars, J. Geophys. Res., 98, 17,049-17,060, 1993.

Webb, R- H., P. T, Pringle, and G. R. Rink, Debris flows in tributaries of the Colorado River in Grand Canyon National Park, Arizona, U.S. Geological Survey Professional Paper 1492,39 pp., 1989.

Wilhelms, D. E., and S. W. Squyres, The martian hemispheric dichotomy may be due to a giant impact, JVa/wre, 309,138-140,1984.

Wise, D. U., M. P. Golombek, and G. E. McGill, Tharsis province of Mars: Geologic sequence, geometry, and a deformation mechanism, Icarus, 38,456-472,1979.

Witbeck, N. E., K. L. Tanaka, and D. H. Scott, Geologic map of the Valles Marineris region. Mars (east half and west hal^, USGS Misc. Inv. Ser. Map 1-2010 (1:2,000,000), 1991. 129

Zimbelman, J. R., R. A. Craddock, and R. Greeley, Geologic map of the MTM -15147 quadrangle, Mangala Valles region of Mars, USGS Misc. Inv. Ser. Map 1-2402 (1:500,000), 1994.

Zimbelman, J. R., D. A. Crown, J. A. Grant, and D. M. Hooper, The Medusae Fossae Formation, Amazonis Planitia, Mars: evaluation of proposed hypotheses of origin, (abstract), iMwor. Planet. Set Con/I, XXVIII, 1623-1624,1997.

Zimbelman, J. R., S. E. H. Sakimoto, and H. Frey, Evidence for a fluvial contribution to the complex story of the Medusae Fossae Formation on Mars, Geol. Soc. Am. An. Meeting, abstract 50423,2000. 130

Figure 1. MOLA shaded relief map of the Tharsis and surrounding regions (courtesy of the MOLA Science Team). Also shown are the newly identified northwestern slope valleys (NSVs) and scarps (s) that may indicate late-stage fluvial activity. 131

kunae at •mi Planun#* Amm •—ler-

\^ical cxaggcralicR 15 OX

Msngala '/==»= S.Domg I Ncih Shslbatans Simud P""" TharsiS Tiu J2 -Jlil NpM sicic of ;^crieal mstcfials

Figure 2. (A) MOLA shaded relief map showing features of interest, including NSVs,

Tharsis rise, Circum-Chiyse outflow channel systems, and centers of tectonic activity

(Anderson et al., 1998; Dohm et al., 1998) interpreted to represent magmatic-driven doming events (UD - Ulysses, ASD - Ascaeus-south; ASWD - Arsia-southwest; CVD - central Valles; SD - Syria), (B) part of the Geologic map of the western equatorial region of Mars (representative map units are shovm-Scott and Tanaka, 1986), (C) generalized geologic cross section (a-a'—transect of A and B; PBS - potential location of basement structures; TMl- Tharsis Montes lavas; MF - Medusae Fossae materials). 133

Figure 3. 3-D topographic projection of the NSVs and surrounding region using

ARCVIEW (MOLA data courtesy of MOLA Science Team). White arrow represents potential Noachian/Early Hesperian catastrophic flooding from source region (Arsia

SW/Syria dome complex). Also shown are centers of tectonic activity (after Anderson et al., 1998; Dohm et al., 1998) interpreted to represent magmatic-driven magmatic doming events (UD - Ulysses dome, ASD - Ascaeus-south dome; ASWD - Arsia-southwest dome; CVD - central Valles dome; SD - Syria dome). 134

Chiyse

/y^raunius /empe SVs y/^ ^-^iKrcuip^fhryse / ' z'' UrSBfiuS OujB^w ChanjT^ ' ^-Tystems / /Arsia S' /Clsffltas^ Thaumasia plateau/^

Wareggo

Argyre

Figure 4. Schematic diagram showing Late Noachian/Early Hesperian Activity. Centers of tectonic activity—Tempe, Uranius, Ceraunius, Arsia SW, Syria, central Valles, Claritas, and Warrego [Anderson et al., 1998; Dohm et al., 1998] are inteipreted to be magmatic-related doming events and, in many cases, associated volcanic events. The formation of NSVs may be partly related to the development of the Arsia SW/Syria dome complex. Also shown is modification along the highland-lowland boundary and Thaumasia plateau (uplift shown by arrows), Coprates rise, Thaumasia highlands, early circum-Chryse outflow channel systems, early Valles Marineris canyon system, Uzboi, fault and fracture system (straight lines), and wrinkle ridge development (wavy lines). 135

0

Figure 5. Smooth chamiel floor (black arrows; at 30 m/pixei) exposed among outcrops and boulders of Medusae Fossae Formation (MFF) materials. The smooth floor does not constrain whether the flow preceded or post-dated MFF, but the lack of any streamlining on knobs within the channel may argue an earlier stage of development. The chaimel gently slopes (0.22°; towards upper left) along the bottom of a broad topographic valley revealed by MOLA data [Zimbelman et al., 2000]. Unusual flow-like features (white arrows) crop out from beneath the MFF deposits on both sides of the channel, and are 136 visible through the MFF cover (right center); these flows have been interpreted as possible peperites [Gregg and Schultz, 1997], implying the flows may have been emplaced within early MFF materials made wet by flow along the chaimeL Area shown is 3.0° to 4.0°N, 140.0° to 141.2°W [U.S.G.S., 1995]. 137

APPENDIX C

COMPARATIVE HYDROGEOMORPHOLOGY OF WET BEAVER CREEK, ARIZONA, AND ABUS VALLIS, MARS.

J. C. Ferris, J. M. Dohm, and T. M. Hare

Manuscript submitted. 138

COMPARATIVE HYDROGEOMORPHOLOGY OF WET BEAVER CREEK,

ARIZONA, AND ABUS VALLIS, MARS.

J. C. Ferris^ J. M. Dohm', and T. M. Hare^; 'Department of Hydrology and Water

Resources, The University of Arizona, Tucson, AZ, 85721, [email protected]:

^U.S. Geological Survey, Flagstaff, AZ.

ABSTRACT

A detailed investigation of Wet Beaver Creek, Arizona, revealed an intriguing amount of geologic, hydrogeologic, and geomoiphic similarities between Wet Beaver

Creek and Abus Vallis, Mars, and its surrounding sapping chaonels. The channel floor of

Wet Beaver Creek is dominated by well-imbricated cobbles and boulders, averaging between a half-foot and two feet in diameter. We determined these boulders were emplaced from multiple flood events, with at least one possessing a peak discharge of

42,000 cubic meters per second (1,500,000 cubic feet per second). More intriguing, we determined the mechanism for this extraordinary flooding to be predominately non- darcian groundwater discharge. This is attributed to a unique set of geologic conditions, particularly; 1) Wet Beaver Creek occurs along the Mogollon rim, a massive scaip and elevation dichotomy caused by the uplift of the Colorado Plateau, 2) Wet Beaver Creek occurs within a region of intersecting, deep-seated basement structure and fault swarms.

We then analyzed Abus Vallis using Mars Orbital Camera (MOC) imagery and Mars

Orbital Laser Altimeter (MOLA) data. This, coupled with previous investigations and geologic maps, allowed us to investigate the paleohydrology of Abus Vallis and similar 139 sapping channels in great detail. We detennined Abus Vallis to be the result of similar hydrogeomorphic processes as Wet Beaver Creek. However, under current martian surface conditions, water is hydrodynamically unstable and Mars is theoretically covered in a planet-wide cryosphere. After examining multiple possible mechanisms for liquid water to be present in such an amoxmt to allow for the formation of groimdwater sapping channels, we devised multiple working hypotheses that included hypersaline aquifers, hydrothermal processes, and global climatic change. Given the similarity in genesis between Wet Beaver Creek and Abus Vallis, one may infer lithologic and hydrologic similarities as well. Wet Beaver Creek is an example of geologic conditions that allow the rapid, sometimes dynamic, creation of sapping channels. With this terrestrial analog, sapping chaimels that occur along the highland-lowland boundary scarp, such as Abus

Vallis, may now also be explained in terms of rapid formation during short-term periods of global climatic change, as well as by hypersaline aquifers and hydrothermal processes. 140

INTRODUCTION

A comparative geomoiphic and paleohydrologic investigation was conducted on

Wet Beaver Creek, Arizona, and Abus Vallis, Mars. In order to better \mderstand late- stage fluvial processes on Mars, Wet Beaver Creek was selected because of its morphometric similarity to sapping channels on Mars (Figures 1 and 2). The availability of high-resolution topographic data from the Mars Orbital Laser Altimeter (MOLA), high-resolution imagery from the Mars Orbital Camera (MOC), and state-of-the-art hydrologic models, provided the ojjportunity to improve upon current imderstanding of martian sapping channel formation (Howard et al., 1988). Previous investigations focused primarily on comparing the sapping channels of the Colorado plateau to those foxmd on Mars, and inferring similar genesis from similar geomorphology (Kochel et al.,

1985; Laity, 1988). Other investigations used flume models with simulated martian surface materials to model sapping channel formation (Howard, 1988; Phillips, 1988).

This comprehensive investigation revealed a dynamic system, unique when compared to tbe sq)ping channels of the Colorado Plateau, providing an additional analog to help explain sapping chamiel formation on Mars.

WET BEAVER CREEK GEOLOGIC SETTING

Wet Beaver Creek is a perennial stream located in central Arizona, east of the

Verde Valley and south of the White Mesa (Figure 1). The regional geology surrounding

Wet Beaver Creek is described here in accordance with the geologic map of Weir et al.

(1989). The regional geology is predominantiy sedimentary rocks capped by Miocene to 141 early Pliocene basaltic lavas. The volcanic rocks are principally medium gray to dark gray, aphanitic to dictytaxitic in texture, and often vesicular. The groxmdmass is generally micrograined to very fine-grained, composed of olivine, plagioclase, and clinopyroxene (Holm, 1994). The lava flows, which generally source at structurally controlled alignments of shield volcanoes, cinder cones, and lava domes in the Mormon

Volcanic Field, form flow units on the southwestern margin of the Colorado Plateau and along the Mogollon lim varying from lO's to lOO's of feet in thickness, and xmconformably overlie the Permian Kaibab Formation and the undivided Toroweap

Formation and Coconino Sandstone. These formations, and the lava flows, chiefly compose the valley walls of Wet Beaver Creek. Wet Beaver Creek is bounded at various reaches by basaltic vent deposits that commonly form small shield volcanoes or eroded cones (i.e., Casner Butte). They consist mainly of lavas, agglomerate, spatter, and relatively small amounts of cinder, and commonly expose dike complexes and plugs.

The headwaters of Wet Beaver Creek expose the Kaibab Formation, a lower

Permian dolomite and limestone. It is yellowish to light-gray, and silty to sandy in texture. Moving downstream, the valley walls are dominated by the undivided Toroweap

Formation and Coconino Sandstone. This unit is comprised of lower Permian quartzose sandstone. It is light to yellowish gray, and very fine to fine-grained in texture. It displays medium to large-scale crossbeds, with rare occurrences of horizontal bedding often near the base of the unit Moving further downstream, the valley walls are composed of the undivided upper and middle parts of the Supai Formation. This unit is composed of sandstone, siltstone, limestone, dolomitic limestone, and minor amounts of 142 conglomerate. In the study area, it is reddish-brown and very fine to fine-grained in texture, often displaying irregular and discontinuous ledges (Figure 3). The stream bed, at the downstream part of the valley, has been mapped and described as Holocene alluvium by Weir et al. (1989), composing unconsolidated clay, silt, sand, and gravel of local derivation. This description is mostly consistent with Holocene alluviimi, which we observed in the neighboring Dry Beaver Creek. However, upon thorough investigation, the distal end of the valley containing Wet Beaver Creek does not contain the mapped

Holocene alluvium. In fact, it is devoid of almost any fine-grained material (Figure 4).

Instead the channel floor is dominated by well-imbricated cobbles and boulders, averaging between a half-foot and two feet in diameter. They are mainly comprised of mid-Miocene to early Pliocene basalts, derived jfrom outside the valley system

(presumably by fluvial-erosional transport), and a small percentage of Permian sandstone and siltstones. This small contribution of wall rock to the channel floor was imexpected and significant, revealing that this system was more dynamic than previously anticipated.

HYDROGEOMORPHOLOGY AND PALEOHYDROLOGY OF WET BEAVER

CREEK

Wet Beaver Creek is best characterized as a simple channel system. It has a drainage basin of 178 km^ (111 mi^ Figure 1). Hie main channel is approximately 15 kilometers (9 miles) long. The average width of the valley is approximately 1200 meters

(4000 feet), and the depth from the top of the canyon is approximately 250 meters (800 feet), yielding a width-to-depth ratio of 5. As previously-mentioned, the channel is 143 predominately composed of Supai Formation sandstone that is capped by Miocene basalts (Weir et al., 1989). Descending into the valley, the tertiary basalts form a near vertical face over a hundred feet thick, then the valley shallows to an approximately 50 degree slope, followed by an abrupt incisement into the main channel. Most of the channel walls are covered in scrub vegetation, interrupted occasionally by debris aprons emanating from the overlying basalt. Outcropping from the margins of the main channel are numerous cobble to sand-size deposits of fining-upward sequences, possibly a result of alluvial processes. The sides of the channel, which are composed of sandstone, display terracing for certain reaches. On the floor of the channel, the perennial stream could well be categorized as "imderfit." The perennial stream tends to the left side of the channel, but just downstream of where Figxu« 4 was taken, it moves laterally across the channel floor to the right side. The channel floor is composed mainly of imbricated subangular to well-rounded large cobbles and boulders (Figure 4). Thirty foot trees grow intermittently on the boulder floor, with some toppled over to reveal their root systems interwoven into the large clasts. The stream is imderlain with the same materials as the channel, with the exception of large depressions in the active stream which fonn deep, placid pools of water. Noticeably absent from the stream and the channel floor is the presence of fines, such as alluvial sands or silts.

In order to better understand Wet Beaver Creek, we also investigated the nearest fluvial system. Dry Beaver Creek. Dry Beaver Creek has a drainage basin of 368 km^

(142 mi^). Unlike Wet Beaver Creek, the floor of Dry Beaver Creek in not composed of imbricated large cobbles and boulders, histead, it is a poorly sorted collection of 144 materials, ranging from large cobble size clasts to fine silts. The channel floor itself is dominated by the presence of alluvial sands and silts. This was initially imexpected, as

Dry Beaver Creek accesses a much larger watershed than Wet Beaver Creek. The peak discharge history of Dry Beaver and Wet Beaver Creeks reveals that Dry Beaver Creek has historically produced floods larger in volimie than those found at Wet Beaver Creek

(Figure 5). These observations raise the question: Why does Wet Beaver Creek have better sorted sediments and show evidence of extraordinary paleoflood events, whereas

Dry Beaver Creek does not? The Weir et al. (1989) geological map, the topographic map

(USGS, 1965), and our field investigations all reveal that the headwaters of Dry Beaver

Creek, while pulling from a larger watershed than Wet Beaver Creek, do not intersect structural faulting that draws from groundwater. This is also seen in Figure 6, which shows a comparison of the monthly average discharge for 1999, a typical year for

Arizona climate. Dry Beaver Creek has no flow for many of the months. When it does flow, it is due to the input of precipitation. Alternatively, Wet Beaver Creek is shown to flow perermially, maintaining a base discharge of approximately 0.21 cubic meters per second (cms), or 7.5 cubic feet per second (cfs). This explains the absence of fines, as any such material that is deposited on the chaimel floor would soon be transported by eolian processes into contact with the perennially active part of the channel and removed.

However, this process, and the historical flood record, cannot explain the existence of the imbricated large cobbles and boulders which compose the channel floor.

We hypothesize that the imbricated large cobbles and boulders resulted fix)m floods of a much greater magnitude than ^^t is recorded in the historical record. The 145 valley was searched for evidence of paleostage indicators (PSIs). The historical floods had left a rich record of tree scars and organic drift, but these floods were far too small to explain our observations. Although there were no slackwater deposits observed, we located a distinct and prevalent hillslope trimline, which truncated the basaltic debris aprons (Figure 7). Trimlines are erosional features caused when floodwaters remove bankside material. Of all possible PSIs, this is considered the least desirable, due to the large imcertainty in water surface height above the trimline (Webb and Jarrett, 2002).

However, we assumed the trimline to represent the maximum peak discharge in order to ensure the paleodischarge estimates were conservative. To determine the paleodischarge, cross-sections were obtained from the U.S. Geological Survey Casner Butte Quadrangle

7.5 minute topographic map (USGS, 1965). We performed a step-backwater analysis using HEC-RAS (U.S. Army Corps of Engineers, 1998; i.e.. Figure 8). The resultant discharge was 42,000 cms (1,500,000 cfs), approximately two orders of magnitude greater than the maximum historical flood (Figure 5). Stream power is an expression for the rate of potential energy expenditure per unit length of chaimel. At the location of

Figure 4, we calculated a stream power value of 636,800 kilograms per second (kg/s), or

1,404,000 pounds per second (Ib/s), su£5cient to entrain the large cobbles and boulders by rolling, sliding, saltation, ccnd as suspended load (Knighton, 1998). Furthermore, the deep pools of water observed next to the terraced sandstone cliffs (Figure 3) may result from eddying and scour of the softer material diiring one of these extraordinary flood events. We sought to confirm the step-backwater model by using another method for estimating paleodischarge based on streambed competency. This methodology is subject 146 to large errors due to inherent problems with correlating the size of entrained boulders to discharge; nevertheless, depending on the mean clast size (boulder, small cobble, etc.) we calculated discharge values between 2000 cms (80,000 cfs) and 7000 cms (250,000 cfs) based on streambed competency.

The imbricated large cobbles and boulders occur in two distinct deposits (Figure 4) and are interpreted to represent at least two extraordinary paleofloods. The larger deposit represents the original emplacement of these materials, with the finer constituents removed by lesser floods typified in the historical record (Figure 5). The smaller deposit represents the occurrence of a second flood, close if not equal in magnitude to the original flood that emplaced this unit. The latter flood was powerful enough to move tiiese boulders, perhaps in the thalweg of the flood channel. In fact, very near wiiere the stream makes a shift from the left to the right side of the channel floor, the boulders have been removed in a similarly transverse pattern. Our hypothesis is this second flood may have been nearly equivalent to the original flood which emplaced the large deposit, but it was unable to produce or sustain the eqmvalent stream power to fully remove flie first deposit As such, it dissected and removed material from the large deposit, forming what has been referred to as the small deposit.

The mechanism for such large floods in a channel that has only three, stubby tributaries and a watershed that is disproportionately small to the channel size is complex.

The drainage basin for Wet Beaver Creek is 286 km^ (111 mi^). For comparison. Dry

Beaver Creek has a drainage basin of 367 km^ (142 mi^; Pope et al., 1998). The mean annual precipitation is 630 mm (24.8 in) and occurs sporadically, the bulk derived fix)m 147 the summer monsoon season and winter rains and snow. However, even the most extreme of hydroclimatic conditions, such as a stalled hmricane or tropical storm, could not have produced a 42,000 cms (1,500,000 cfs) flood for the Wet Beaver Creek. This was determined using a variation of the Rational Method. For purposes of simplicity and conservative estimation, we assiuned an infiltration rate of zero, when in fact the groundmass in highly permeable. The appearance of fissures supports this observation

(Figure 9). We also assumed that the transfer of surface water from the entire drainage basin to the channel was instantaneous, although that is impossible. Finally, we assumed that the entire drainage basin was covered in uniform rainfall, an unlikely assumption given the size of the watershed. Using the equation

Qpk = /*As

Where Qpk is the peak discharge, I is the rainfall intensity, and As is the saturated area, assumed here to represent the entire drainage basin. Even with all the conservative assumptions, a peak discharge of 42,000 cms (1,500,000 cfs) would require a rainfall intensity of 536 mm (21.1 in) per hour. Given the assumptions made here, the actual required rainfall intensity value is probably much higher. For comparison, the highest precipitation value ever recorded was at Belouve, on La Reimion Island, a French territory in the Indian Ocean. Over a twelve hour period, they received 1350 mm (53 in) of rain O'erth - Local and World Weather, 1998). That represents a mean rainfall intensity of 112 mm (4.42 in) per hour. The amount required at Wet Beaver Creek, a vastly different hydroclimatic environment than a small island in the Indian Ocean, is almost five times as much. Clearly, surface precipitation and runoff is not enough to 148

explain the paleoflood evidence. We propose that these extraordinary floods are not

solely the result of surface water flow, or that sxnface water is even (directly) the major

contributor. Instead, we propose that these extraordinary floods have resulted from the

same processes that have kept the stream perennial and led to the growth of the valley:

groimdwater discharge.

The hydrogeomorphology of Wet Beaver Creek is different from traditional

groundwater sapping channels, typified on the Colorado Plateau (Howard et al., 1988).

Groundwater sapping is a generic term for the weathering and erosion of soils and rocks

by emerging groimdwater. Groundwater sapping must at least partially involve

inteigranular flow, as opposed to the channelized throughflow involved in piping. The

Colorado Plateau is underlaid by cake-layer sequences of Mesozoic and Paleozoic sediments, which are primarily shales and sandstones, with some interbedded formations of limestone, gypsum, and/or more soluble evaporites. The bedding is primarily flat, but is locally faulted or warped into monoclines, domes, and basins. The sandstones weather easily due to high porosity, permeability, and weak sedimentation. The mechanisms by which they erode include grain-by-grain surface removal, separation along bedding plains, and crumbling. The sandstone units on the Colorado Plateau are exposed as bare rock slopes except where they are mantled with eolian sands or alluvium. Exceptions to this are areas of very low relief, such as the backslopes of gently dij^ing sandstone cuestas, vdiich may be mantled with poorly horizonated soils and scrubby vegetation. The re^onal joint pattern exerts considerable control on the groundwater drainage pattern of the Navajo Sandstone. The fracturing increases the overall permeability of the plateau 149 surface, and increases the ability of precipitation to contribute to the water table. The fiactures act at depth to increase the transmissivily of the bedrock, while laterally flowing groimdwater exploits the major joints. As a result of this, the water emerges at various seepage points along cliff faces, while the canyons migrate headward along joint trends.

The sapping process undennines the cliff faces and causes the collapse of massive sandstone slabs. The exfoliation joints on Ihe Colorado Plateau are best developed in the more massive imits, which includes the Navajo Sandstone. It is in this unit that exfoliation joints appear to result jfrom pressure release from canyon cutting. These joints are developed parallel to the sidewalls, and it is the collapse of these massive slabs caused by exfoliation jointing that result in the high and steep valley sidewalls. The large slabs shatter on impact and are quickly reduced to fines by active weathering processes.

Theatre-headed valleys result from groundwater outflow and sapping processes. These valleys develop where large exposures of gently dipping and highly fractured sandstone allow high rainfall infiltration rates (Laity, 1988). The groundwater encounters a permeable boundary, and moves laterally down the hydraulic gradient through intergranular pores and firactures. The flow converges at valley heads, and emerges in concentrated zones of seepage. These zones allow for small-scale erosional processes to slowly reduce the support of steep cliffs, which will ultimately bring forth their collapse.

These valleys grow headward by successive slab failure, aided by processes that break down the resulting debris and carry it away. Initially, growth in the main canyon proceeds more quickly than growth in the tributary canyons, due to its larger subsurface drainage area. In the advanced stages of formation, however, the headward growth rates 150 diminish as the system enlarges and the drainage area lessens. This helps create conditions in which the lateral retreat by sidewall seepage equals the rate of headward retreat, and the valley starts to widen (Laity, 1988).

The hydrogeomorphic differences between Wet Beaver Creek and the sapping channels of the Colorado Plateau is due to the modification of the valley walls and slope by the presence of surface runoff down the sides of the valley due to intense seasonal floodmg, something the sapping channels of the Colorado Plateau lack due to geography and hydroclimatic conditions. The gross morphology, however, is consistent with sapping-dominated processes. Wet Beaver Creek has only three stubby tributaries, most likely the result of surface water exploitation of knick points. The main channel meanders back, carving its way aroxmd obstacles (i.e., Casner Butte) and terminating in a quasi-theater headed morphology (Figure 10). Wet Beaver Creek continues upstream, but only as a small, ribbon channel that exists as a result of surface water discharge.

However, the sapping channels of the Colorado Plateau do not exhibit the extraordinary level of flooding inferred at Wet Beaver Creek. We hypothesize tiiis is because Wet Beaver Creek is encompassed by a unique set of geologic conditions. It occurs along the Mogollon rim, a massive scarp and elevation dichotomy caused by the

Colorado Plateau uplift The changing stress regimes through time, including the

Colorado Plateau uplift, have resulted in the generation of numerous fault swarms. Weir et al. (1989) notes at least 16 small normal faults running transverse to Wet Beaver Creek.

Also shown is a large, curvilinear fault that also shows up pronainently on Side-Looking

Aperture RADAR (SLAR, Figure 1). The downdropped side is very close to the 151 headwaters of Wet Beaver Creek, as well as marks a distinct morphological shift for

West Clear Creek to the south. These, however, are merely the faults which have surface expression. Detailed geologic investigations in the region and to the north (Dohm, 1995;

Bills et al., 2000) reveal north-, northwest-, and northeast-trending fault systems, including deep-seated basement structures. The deep-seated basement structures could represent long-lived features that have been reactivated numerous times in response to changing stress regimes (Shoemaker et al., 1978; Dohm 1995). Where these systems interact, unique hydrologic circumstances tend to occur (i.e., Stoneman Lake; Dohm

1995). Based on these investigations, we believe the headwaters of Wet Beaver Creek occur within a region of interacting tectonic structures. While Wet Beaver Creek may not possess an extensive surface watershed, it may have access to a tremendous supply of groundwater. The regional tilt imposed by the Mogollon rim and associated uplift have created conditions that lead to regional grovmdwater flow towards Wet Beaver Creek.

The regional ground surface is highly porous and permeable, with visible fissures occurring within the groxmdmass (Figure 9). This allows for the rapid infiltration of rainfall and snowmelt into the aquifer system. The extensive faulting may allow for non- darcian flow to occur, provided the joint aperture is large enough.

The paleodischarge estimates for Wet Beaver Creek are solid, with the clear PSIs and multiple estimation techniques resulting in high confidence for the peak discharge values. The occurrence of pervasive, deqj-seated basemrat structural fabric at the headwaters of a stream recording extraordinary floods that cannot result from its surface catchment is very unlikely to be coincidence. We hypothesize that these floods resulted 152 j&om a combination of hydroclimatic and hydrogeologic processes; the only question remaining is the mechanism that causes these floods to occur. Without being able to confirm that one mechanism is more probable than another, we have chosen to use a multiple working hypothesis vmtil more data becomes available. The best-candidate mechanisms are as follows:

The Hydroclimatic and Hydrogeolo^ Collaboration Hypothesis.

This hypothesis states that an extreme hydroclimatic event, such as a stalled hurricane or tropical storm, drops a large amoimt of precipitation onto the southem margin of the Colorado Plateau and Mogollon rim in a short amount of time. While the potential contribution of the surface watershed has already been discussed, this hypothesis states that the groundwater reservoir available to Wet Beaver Creek is much larger. As the rain hits the ground, it is rapidly infiltrated by the sinface fissures and permeable groundmass. It then begins to accxmiulate in the massive deep-seated fi:acture network which focuses the quickly infiltrated groundwater into the head of Wet Beaver

Creek at non-darcian velocities. The problem with this hypothesis is that the surface infiltration rate, density of joints, and size of the jointed network must be very large in order to provide enough groimdwater to help feed a flood that will peak at 42,000 cms

(1,500,000 cfs). The effective permeability can be determined using an equation devised by Snow (1968) that treats a cubic system of like fiactures as an isotropic system, a reasonable approximation for our purposes. The permeability, ^ is equal to

Nb^/6 153

Where N equals the number of joints per unit distance across the rock face, and b equals the aperture of the jomts. It is apparent that there exists a continuum of values for N and b that would result in the required permeability for this hypothesis to be valid. However, these values lie in a range that is highly improbable. When you eliminate the isotropic system assumptions, and allow for the possibility of the deep-seated fracture network to be focusing the groundwater at the head of Wet Beaver Creek, the possibility becomes far more probable of the resulting discharge being orders of magnitude larger than the surface water catchment would allow.

The remaining two hypotheses both seek to circumvent the need for rapid infiltration and regional focusing of groundwater by instead providing a "stopper" mechanism which allows groxmdwater to accumulate in the system. Eventually, an event, perhaps long-time accumulation or an extremely wet season forces the water table to reach a threshold that allows a catastrophic breach at the head of Wet Beaver Creek from which the groundwater in the fracture network rapidly springs forth.

The Catastrophic Cottapse Hypothesis.

This hypothesis states that Wet Beaver Creek originated as a simple surface drainage stream caused by kniclqwiut exploitation. As it ate back into the MogoUon rim, it intersected the fault swarms and their related deep-seated fabric. At this point, there was a catastrophic release of the groundwater stored in the basin above the rim. A massive flood ensued, and the non-daician velocity of the water within the fracture network caused sevore erosion at the head of the channel. Eventually, decreased hydraulic head allowed for coll^se at the head of the channel. This sealed the fractures 154 near the channel head and once again prevented their movement through Wet Beaver

Creek. Over time, the channel either eats back via surface water erosion and/or the hydraulic head reaches a height in which the hydrostatic pressure can forcibly cause another cycle of catastrophic release through the head of Wet Beaver Creek. This hypothesis is supported by evidence of severe collapse at the head of Wet Beaver Creek

(Figure 10).

The PrecipUate Containment Hypothesis.

This hypothesis poses that the aquifer from which Wet Beaver Creek draws is rich in minerals and salts (Bills et al., 2000). As a catastrophic release occurs, the changes in pressure allow for these chemicals to precipitate out of solution. As they do, they reduce the aperture of the joints, and eventually seal them shut Eventually, an increased hydraulic head, or perhaps fault reactivation from earthquakes, which are historically common to this region (Brumbaugh, 1986), forces the precipitate contaiimient to fail.

There are two serious problems with this hypothesis. First, as the aperture begins to narrow due to precipitate on the joint surfaces, the velocity would increase due to the continuity equation. This would, in turn, increase the erosive power at these points.

Second, it is hard to imagine a chemical precipitate that could hold back groundwater flowing at velocities necessary to siqjport a flood with a peak discharge of 42,000 cms

(1,500,000 cfs). PeAaps precipitates play a role in sealing off the groundwater discharge, but it seems imlikely that they are solely responsible.

It is possible that all of these hypotheses play a role in the extraordinary flooding documented at Wet Beaver Creek, or that none of them do. The only viable alternative to 155 hydrogeomorphic processes that we can conceive is a hyperconcentrated debris flow.

That might be able to achieve the same geomorphic expression with considerably less water, although other indicators of a massive debris flow were not observed. It will take considerable more study to determine the exact mechanism of these imusual events.

However, there are locations on Mars where the concept of extraordinary or catastrophic groundwater discharge is routinely invoked to explain the observed hydrogeomoiphology.

There are well-documented locations where floods of immense magnitude (approximately

10^ to 10^ m^/s) have been hypothesized to occur as the result of groimdwater discharge

(Robinson and Tanaka, 1990). These locations have also been hypothesized to possess a highly pemieable surface. Similar to the topographic relief of the MogoUon rim, gigantic valleys (Dohm et al., 2001) border a massive scarp and elevation dichotomy at which nimierous sapping channels are observed that have morphology similar to Wet Beaver

Creek (Figure 2).

ABUS VALLIS GEOLOGIC SETTING

Martian geologic time is divided into three periods, the Noachian, the Hesperian, and the Amazonian. Absolute dates for Mars do not exist, so these time periods are divided on the basis of impact crater densities. The oldest is the Noachian Period, which consists of ancient crustal rocks formed during the period of late heavy bombardment, producing a high density of impact craters. The Hesperian Period records extensive volcanism, including the emplacement of wrinkle ridged plains materials. In addition, widespread resurfacing of Early Hesperian and older materials, which includes the 156 deposition of colluvial and fluvial deposits, is observed during the Hesperian, as well as outflow channel development at Mangala Valles (located to the southwest of the Abus

Vallis) and at the circum-Chryse outflow channel systems located to the north-northeast of Valles Marineris. The youngest period is the Amazonian, which includes the emplacement of associated lavas and aureole deposits of Tharsis Montes and Olympus

Mons, and late-stage outflow channel development at Mangala Valles.

The channel of interest is Abus Vallis, a sapping channel that can be found at the southern margin of the three northwestem slope valleys (NSVs; Dohm et al., 2001). It occurs transverse to the boundary between relatively yoimg, uncratered materials of the lowlands and the highly cratered, ancient materials of the highlands. The lowland materials are interpreted to have been emplaced by both subaerial and subaqueous processes, which include eolian, volcanic, mass-wasting, fluvial, lacustrine, and marine

(Scott and Tanaka, 1986; Greeley and Guest, 1987; Parker et al., 1993; Scott et al., 1995).

The ancient highlands consist of materials that are inteipreted to include lava flows, impact breccias, and eolian, fluvial, and colluvial deposits (Scott and Tanaka, 1986;

Greeley and Guest, 1987). Additionally, Kargel et al. (1995) has inferred various materials to result from glacial processes.

The regional geology surrounding Abus Vallis is described here in accordance with the Chapman et al. (1989) geologic map. The valley trends northward with its headwaters to the south (Figure 2). All of the sapping channels in this region trend northward, most likely due to structural control. The west and northeast sides of the valley is Hesperian-age intercrater plains material. This material, which is interpreted to 157 be composed of intercalated basaltic flows and volcaniclastic deposits, mantles the ridged plateau materials that Abns Vallis dissects. The head of the sapping channel is surrounded by three dififerent units. To the southwest is the Noachian-age ridged plateau material. It is interpreted to consist of early martian crust that has been highly-modified by heavy impact bombardment and erosion by mass-wasting, eolian, and fluvial processes. Directly south, the head of the valley terminates in the undivided Amazonian and Hesperian-age old lobate plains material. This is interpreted to be composed of north-flowing basaltic lavas extruding from vents to the south. East of the sapping channel head, and continuing northward for half the length of Abus Vallis, is Member 3 of the Tharsis Monies Foimation. It is interpreted to consist of basaltic lavas with well- defined flow fi-onts and leveed lava channels associated with Tharsis volcanism. The undivided Amazonian and Hesperian-age wall materials are admittedly poor to distinguish. They are interpreted to be aprons probably consisting of sequences of lava flows, crater ejecta, and fluvial and eolian materials. Finally, the undivided Amazonian and Hesperian-age material of the theatre-headed channels, which compose the floor of

Abus Vallis, are interpreted to be yoimg fluvial deposits resulting firom groundwater sapping. Abus Vallis deposits a large lobe of this material at its distal end, which partly buries the old lobate plains material.

HYDROGEOMORPHOLOGY AND PALEOHYDROLOGY OF ABUS VALLIS

Nrimerous sapping channels occur at the highland-lowland boundary, east of

Mangala Valles and its branches. Moving east, along the dichotomy, are the eight 158 sapping channel systems in our study region: Tinia Valles, Taus Vallis, Dubis Vallis,

Heraius Vallis, Abus Vallis, Senus Valles, Isara Valles, and Munda Vallis (Figure 2).

Most, if not all, would make excellent candidates for Wet Beaver Creek analogs. While

Taus Vallis is more comparable in size to Wet Beaver Creek, and Dubis Vallis more representative of its plan view morphology, Abus Vallis was selected because it is the most well-developed of the sapping channels, and therefore, more representative of the sapping processes behind their formation. Also, Abus Vallis is the only sapping channel large enough to confidently utilize the Mars Orbital Laser Altimeter (MOLA) data-

Previous investigations of the region (Chapman et al., 1989) interpreted the small channels to have formed dining three separate channeling events ranging Jfrom the Late

Hesperian to Early Amazonian. The small channels are divided into three morphologically different types: the Mangala Valles branches, the theatre-headed channels (also referred to here as the sapping channels), and the ribbon charmels. The latter two are classified by their width-to-depth ratios. The s^ping charmels are noted to range fi-om 100 to 1,500 meters in depth (300 to 5000 feet), and firom 1 to 3 kilometers in width (0.6 to 2 miles). This results in average sapping channel width-to-depth ratio of 11.

It should be noted; however, that recent advancements in available data, particularly

MOLA and Mars Orbital Camera (MOC) imagery, may lead to revision of this regional average. Although many are ascribed to lava channeling, two of the Early Amazonian- age ribbon charmels display characteristics that indicate a fluvial origin. Chapman et al.

(1989) raise the possibility that ribbon charmels and sapping channels are morphologic end members of a continuous process. 159

We analyzed Abus Vallis using MOC imagery and MOLA transects. The MOC data was limited; the available imagery showed only the downstream end of Abus Vallis

(Figure 2). Nevertheless, those images showed a featureless channel floor, surprisingly devoid of impact craters. This may be the result of eolian processes, chaimel wall collapse, or perhaps fluvial processes. If so, it may indicate that these processes were active much later than previously considered. Abus Vallis contains an inner ribbon channel, suggesting the occurrence of downcutting at a later date than the sapping channel formation, and/or perhaps a change in the availability of groundwater for fluvial erosion. The MOLA transects, plus a profile running the length of the chaimel floor, allowed greater accuracy to be attained for the channel dimensions. The chaimel meanders 70 kilometers (40 miles), with an average width of approximately 2500 meters

(8200 feet), and an average depth of approximately 200 meters (700 feet). This yields a width-to-depth ratio of approximately 12.5. More importantly, the channel profile allowed for an extremely accurate measurement of the channel slope, which was determined to be 0.0002.

This new data, in conjunction with previous geologic mapping of the region, allows us to reconstruct the paleohydrology of Abus Vallis. The shorter sapping channels in the vicinity of Abus Vallis display a steeper slope, based on photoclinometric interpretation (Chapman et al., 1989). This, coupled with their smaller length, is characteristic of comparatively less-developed sapping channels. This implies that Abtis

Vallis has been active longer than any other sapping channels in the region. Also, a promontory of Hesperian intercrater plains material prevented the inundation of the 160 outlets of Abus Vallis and Senus Valles by Amazonian-age lobate plains material, interpreted to be the youngest basaltic lava flow in the region (Figure 2). This allows a comparison between two spatially close sapping channel systems. The mouths of Senus

Valles, two sapping channels roughly half the size of Abus Vallis, are inimdated by undivided Amazonian and Hesperian-age basaltic lava flows. However, lobes of material extruded from Abus Vallis, interpreted to be fan deposits formed by fluvial processes, superimpose the undivided Amazonian and Hesperian-age basaltic lava flows. This implies the longest active sapping channel is also more recently active than its contemporaries to the east.

The reasons for Abus Vallis being more-developed and the most recently active sapping channel are numerous. The dominant structure in the region is the highland- lowland boxmdary scarp. Chapman et al. (1989) notes that crosscutting relationships imply that the scarp formed during Late Hesperian time, and that steeper slopes and greater heights of the scarp correlate to greater depths of channel erosion. The faulting of the highland-lowland boundary scarp and channel erosion is most likely interrelated.

This type of event has been modeled in a runoff flmne, where the surface was tilted to simulate base level drop along a fatilt. In these models, sapping chaimel formation occurred immediately and mimicked the observed morphology (Howard, 1988; Phillips,

1988). Aside from the highland-lowland boundary scarp, the regional trend of the structure mimics the northern trend of the sapping channels, as well as the comparatively gigantic Mangala Valles, suggesting they form conduits for water and other volatiles.

Although this provides a mechanism for the formation of these features, in and of itself, it 161 does not explain their formation. This is becaiise under current pressures and temperatures, any water at the surface and below should be frozen as ground-ice, down to extensive depths where the geothennal gradient allows the phase shift into water. So, given the existence of this extensive martian ciyosphere, how did these sapping channels form? To explain the temperature and pressure conundrum, we devised a multiple working hypothesis. While no possibilities can be categorically ruled out, serious investigation has allowed us to determine three of the most likely hypotheses.

Hypothesis 1: Hypersaline groundwater

One explanation is that martian groimdwater is hypersaline, and can exist in fluid form under current martian surface pressure and temperatures (Knauth et al., 2000). This allows for the occurrence of groundwater sapping, but it does not explain why these channels are not more-developed, as it can be assumed they would have been forming since the Late Hesperian. It may be that the groundwater present was exhausted to the point that the hydraulic head no longer drove volatiles along the existing structural controls, in the du«ction of the sapping channels. However, the large spacing between the channels, and their relatively poor-development seems inadequate to drain an area the size of our study region. It could be that the region was uniformly groundwater deficient to begin with, but if so, groundwater sapping would probably never have initiated.

However, Knauth et al. (2000) has proposed that due to various processes, the salinity of tiie near-surface groundwater may vary. It may be tiiat tiiese groundwater sapping channels represented pockets of groundwater that were saline enou^ to remain liquid at current martian temperature and pressure. Once these pockets were exhausted, the 162 growth of the groundwater sapping channels would cease. A iBnal consideration is that the increased salinity would greatly increase the viscosity of the groundwater. It remains uncertain whether, given the low martian gravity (3.82 m/s^), a hi^ hydraulic head could have moved the viscous mineral-enriched groundwater.

Hypothesis 2: Hydrothermal Activity

Another hypothesis is that the extensive volcanism to the east of Abus Vallis may have melted a significant amount of groimd-ice, and drove the volatiles west. They would then intercept the prevailing lines of structure and exploit them at the hi^and- lowland boundary scarp, creating the sapping channels. However, it is unlikely that lava flows, even of the extent of the Tharsis Montes Formation, coiild convect enough heat into the ground to create a hot, large aquifer whose flow alone would have to contdn enough additional heat to melt, and sustain the melt, of all the ground-ice to the west.

Furthermore, the donainate north-trending tectonic structure forms a geographic barrier south and southeast of Abus Vallis, which may also reflect an underlying groundwater divide or aquiclude. The pervasive northward faulting may make it difficult for water to migrate far enough east to melt and supply such sapping channels as Tinia Valles, especially considering the strong nordiward gradient in the area.

Alternatively, the magmatic activity associated with the development of the

Tharsis Montes shield volcanoes may have emplaced intrusives, such as laccoliths and dike swarms (McKenae and Nimmo, 1999), or peiiiaps even a large magma chamber beneath the surface of our study area. This could provide ample heat for the conversion of the existing ground-ice into groundwater. The formation of groundwater sapping 163 channels would contmue until the emplaced magmatic heat and/or the regional groundwater supply was exhausted. It is worth noting that this could explain the differential development among the sapping channels; certain areas may have been exposed to geothermal heating over longer periods of time, or the chamber itself could have been replenished to produce multiple region events.

Hypothesis 3: Climatic Change

Numerous researchers have proposed climatic change as an explanation for fluvial features on Mars that postdate the Early Noachian. A favored hypothesis involves changes in obliquity, which would tilt the planet so that certain regions receive more sunlight and heat This theory has recently received support from experiments in complexity theory \^iiich show martian obliquity to display chaotic properties (Touma and Wisdom, 1993). It is difficult, however, to imagine any change in obliquity that would significantly warm the equatorial regions where Abus Vallis is located. The effects of a change in obliquity would have their greatest effects in the upper latitudes.

The only viable explanation for increasing temperatures in equatorial regions would then have to be a planet-wide increase in temperature. This has been proposed by Baker et al.

(1991, 2000) to result from extensive volcanic outgassing, which increases the density and composition of the martian atmosphere by increasing the amounts of carbon dioxide, methane, and other greenhouse gasses. Interestingly, the Late Hesperian, the time when these s^ping channels are believed to have formed, was also a time of intense volcanic activity (Anderson, 1998; Dohm, 2001a). The third member of the Tharsis Montes

Formation, a basaltic lava flow of undivided Amazonian and Hesperian-age, borders the 164 east, upstream half of Abus Vallis. It is highly unlikely that the flow would simply stop at its meandrous edges, and the MOLA data shows no evidence for infilling, so it reasonable to assume the channel eroded back half its length after the emplacement of the third member of the Tharsis Montes Formation, eroding into the less resistant materials of the intercrater plains materials. The fact it so precisely follows the border of the basaltic lava flow, avoiding it and preferentially eroding the intercrater plains material, calls into question the composition of the ridged plateau material which the intercrater plains material mantles. This is significant, as the ridged plateau material is a remnant of the early martian crust. We hypothesize that the climatic change caused by Late Hesperian volcanism allowed the atmosphere to thicken, and the surface of the planet to wami enough to allow near surface ground-ice to melt into a large groimdwater reservoir, which allowed sapping channel formation as the highland-lowland boundary scarp developed.

This does not rule out any potential contribution fi-om magmatic activity. The climatic change event, dubbed "MEGAOUTFLO" by Baker et al. (2000), is proposed to last between 1,000 and 100,000 years. This would imply that sapping channel formation would have been relatively rapid, and that the extreme length and extensive development of Abus Vallis must result firom either a differential in material strength, a preferenti^ location to prevailing structures, or that Abus Vallis was formed by multiple events. An additional advantage to invoking climatic change as an explanation for sapping channel formation is that it can also explain the decidedly non-sapping ribbon channels, one of which moves upstream firom Abus Vallis' sapping head and another which lines the 165 channel itself. Weak precipitation or snowmelt could explain the ribbon channels that display characteristically fluvial features in this region.

DISCUSSION

A comparison of Wet Beaver Creek, Arizona, and Abus Vallis, Mars, has revealed a surprising amount of geomorphologic similarities, which may imply similar genesis if equifinality is not a concern. Wet Beaver Creek resulted from the uplift of the

Colorado Plateau and the subsequent creation of the Mogollon Rim. We hypothesize groundwater flow through this massive scarp began to exploit areas of permeability and due to the interfingering of at least two distinct fault swarms. Wet Beaver Creek began to erode headward, assisted by precipitation. We further hypothesize this quasi-sapping channel displays evidence of at least two extraordinary floods, most likely due to hydrogeologic processes. ITiese episodes of intense flooding, iq addition to the local geology (i.e., a basalt flow overlying fractured sandstone), may have allowed for the rapid development of Wet Beaver Creek (approximately 5 to 10 Ma). Similar to Wet

Beaver Creek, Abus Vallis has been hypothesized to result from a combination of the formation of the highland-lowland boxmdary scarp, and the occxra-ence of a large area of liquid groundwater from which the sapping channel could draw upon. There is a regional, deep-seated system of northward trending structures which undoubtedly assisted in the growHi of the channel. Although Abus Vallis appears more developed than Wet Beaver

Creek (i.e., the channel is longer and possesses a much gentler slope), this may be due to the erodibility of the material and the presence of that additional north-trending structure. 166

Similar to Wet Beaver Creek, the material which Abus Vallis incises into is overlain with

basalts (the intercrater plains material). What lies beneath is early martian crust, assumed

to be basaltic and modified by mass-wasting, eolian, and fluvial processes. We propose

that this material must be similar to that of Wet Beaver Creek, that is, the basalt contains

layers of sandstone, siltstone, and conglomerate, or perhaps is entirely composed of these

materials. The greater length and development of Abus Vallis also may be due to the

lower martian gravity, which allows water to entrain more and heavier materials by

rolling, sliding, saltation, and as suspended load than it could on Earth. Alternatively, the

material that composes the channel could be even more erodible than its terrestrial

coimterpart, perhaps highly-fiactured and extremely loosely consolidated sandstone.

Given the similarity in genesis between Wet Beaver Creek and Abus Vallis, we might

infer lithologic and hydrologic similarities as well. Wet Beaver Creek may I)© an

example of geologic conditions that allow the rapid, sometimes dynamic, creation of

sapping channels. Sapping channels occurring along the highland-lowland botmdary

scarp, such as Abus Vallis, may represent long-term formation due to a hypersaline aquifer or long-lived hydrodynamic stability due to hydrothermal processes, or a combination of these processes, and may also include climatic change. In addition. Wet

Beaver Creek suggests that groundwater saj^ing processes can become dynamic imder certain circumstances, allowing for the genesis of sapping channels even if periods of climatic change are comparatively brief. If this hypothesis is correct, then Abus Vallis and its contemporaries can be used to understand and constrain martian paleoclimatology. 167

CONCLUSIONS

Wet Beaver Creek is a simple channel system that results from a complex set of

geological and hydrological conditions. It is hydrogeomorphologically similar to Abus

Vallis, and most likely, the sapping channels which occur in the region of Abus Vallis.

Rxiling out equifinality, we concluded that similar processes must be at work in the formation of Abus Vallis. However, the problem remains of the theorized global cryosphere preventing sapping channel development on Mars. We devised a multiple

working hypothesis for liquid water in quantities large enough to support the observed

groundwater sapping, which included hypersaline aquifers, hydrothermal processes, global climatic change, or any combination of the above. Although hypersaline aquifers and hydrothermal processes have been proposed by previous investigators (i.e., Knauth et al., 2000; Howard et al., 1988), our hypotheses explored the ramification of these processes to explain the formation of structurally-controlled groundwater sapping channels in the region east of Mangala Valles. More importantly, using the dynamic

behavior observed at Wet Beaver Creek as an analog, we introduce the idea that these sapping channels on the highland-lowland boimdary scarp could be rapid in their formation, allowing for short-term climatic change events to be a possible progenitor. If

so, these unique s^ping channels may reflect a Late Hesperian and younger climatic change and can be used in similar environs to constrain and better understand martian climatic change. 168

ACKNOWLEDGEMENTS

We thank Jon Pelletier and Katherine Hirschboeck for their ideas and insightful criticism. 169

REFERENCES

Anderson, R. C., M. P. Golombek, B. J. Franklin, K. L. Tanaka, J. M. Dohm, J. H. Lias, and B. Peer, CentCTS of tectonic activity through time for the western hemisphere of Mars, Lunar Planet. Sci. Conf. [CD-ROM], XXDC, abstract 1881, 1998.

Baker, V. R., R. G. Strom, V. C. Guiick, J. S. Kargel, G. Komatsu, and V. S. Kale, Ancient oceans, ice sheets and the hydrological cycle on Mars, Nature, 352, 589-594, 1991.

Baker, V. R., R. G. Strom, J. M. Dohm, V. C. Guiick, J. S. Kargel, G. Komatsu, G. G. Ori, and J. W. Jr. Rice, Mars' Oceanus Borealis, ancient glaciers, and the MEGAOUTFLO hypothesis, Lunar Planet. Sci. Corf. [CD-ROM], XXXI, abstract 1863,2000.

Bills, D. J., M. Truini, M. E. Flynn, H. A. Pierce, R. D. Catchings, and M. J. Rymer, Hydrogeology of the regional aquifer near Flagstaff, Arizona, 1994-97, TJSGS Water- Resources Investigations Report 00-4122,143 pp., 2000.

Brumbaugh, D. S., Historical seismicity in northern Arizona and some tectonic implications {tabs), American Geophysical Union, EOS, v. 67, no. 44, p. 1226,1986.

Chapman, M. G., H. Masursky, and A. L. Dial, Jr., Geologic maps of science study site lA, east Mangala Valles, Mars, USGS Misc. Inv. Ser. I-Map 1962 (1:500,000), 1989.

Dohm, J. M., Origin of Stoneman Lake, and volcano-tectonic relations of Mormon and San Francisco volcanic fields, Arizona, Northern Arizona University, M.S. thesis, 101 pp., 1995.

Dohm, J. M., R. C. Anderson, V. R. Baker, J. C. Ferris, L. P. Rudd, T. M. Hare, J. W. Jr. Rice, R-R. Casavant, R. G. Strom, J.R., Zimbehnan, and D. H. Scott, Latent activity for western Tharsis, Mars: significant flood record exposed, J. Geophys. Res., 106, p. 12,301-12,314,2001.

Dohm, J. M., J. C. Ferris, V. R. Baker, R. C. Anderson, T. M. Hare, R. G. Strom, N. G. Barlow, K. L. Tanaka, J. E. Klonaszewski, D. H. Scott, Ancient Drainage Basin of 170

the Tharsis region. Mars: Potential source for outflow channel systems and putative oceans or paleolakes, J. GeopJ^s. Res., 106, p. 32,943-32,958,2001a.

Greeley, R, and J. E. Guest, Geologic map of the eastern equatorial region of Mars, USGSMisc. Inv. Ser. MapI-1802B (1:15,000,000), 1987.

Holm, R. F., Geologic Map of the Mormon Lake quadrangle, Coconino County, Arizona, Arizona Geological Survey Contributed Map CM-94-C (1:24,000), 1994.

Howard, A. D., Groundwater sapping experiments and modeling. In Sapping Features of the Colorado Plateau: A Comparative Planetary Geology Field Guide, Howard, Kochel, and Holt, ed. NASA Special Paper #491, p. 71-83,1988.

Howard, A. D., R. C. Kochel, and Holt (editors). Sapping Features of the Colorado Plateau: A Comparative Planetary Geology Field Guide, HASA Special Paper #491, 1988.

Kargel, J. S., V. R. Baker, J. E. Beget, J. F. Lockwood, T. L. Pewe, J. S. Shaw, and R. G. Strom, Evidence of ancient continental glaciation in the martian northern plains, J. Geophys. Res., 100, p. 5,351-5,368, 1995.

Knauth, L., P., D, M. Burt, and S. Klonowski, Dense, eurtectic, valley-fomiing, intermediate latitude (DEVIL) brines on Mars, Geol. Soc. America Abstracts with Programs, v. 32, no. 7, p. A303,2000.

Knighton, A. D., Fluvial Forms and Processes: A New Perspective, London, Arnold Publishing, p. 105-141,1998.

Kochel, R. C., Howard, A. D,, and McLane, C. F., Channel networks developed by groundwater sapping in fine-grained sediments: analogs to some martian valleys. In Models in Geomorphology, M. J. Woldenburg, ed. Boston, Allen and Unwin, p. 313- 341,1985.

Laity, J. E., The role of groundwater sapping in valley evolution on the Colorado Plateau, In Sipping Features of the Colorado Plateau: A Comparative Planetary Geology 171

Field Guide, Howard, Kochel, and Holt, ed. NASA Special Paper #491, p. 63-69, 1988.

McKenzie, D., and F. Ninuno, The generation of Martian floods by the melting of ground ice above dykes. Nature, 397,231-233, 1999.

Parker, T. J., D. S. Gorsline, R. S. Saunders, D. C. Pieri, and D. M. Schneeberger, Coastal geomorphology of the martian northern plains, J. Geophys. Res., 98, p. 11,061-11,078, 1993.

Perth - Local and World Weather, World Rainfall Extremes, http://members.iinetnet.au/ ~jacob/worldrf.html, 1998.

Phillips, L., The effect of slope on experimental drainage patterns: Possible applications on Mars, In Sapping Features of the Colorado Plateau: A Comparative Planetary Geology Field Guide, Howard, Kochel, and Holt, ed. NASA Special Paper #491, p. 107,1988.

Pope G. L., P. D. Rigas, and C. F. Smith, Statistical Summaries of Streamflow Data and Characteristics of Drainage Basins for Selected Streamflow-Gaging Stations in Arizona Through Water Year 1996, USGS Water-Resources Investigations Report 98-4225,901^^., 1998.

Robinson, M. S., and K. L. Tanaka, Magnitude of a catastrophic flood event at Kasei Valles, Mars, Geology, 18,902-905,1990.

Scott, D. H., and K. L. Tanaka, Geologic map of the western equatorial region of Mars, USGS Misc. Inv. Ser. MapI-1802-A (1:15,000,000), 1986.

Scott, D. H., J. M. Dohm, and J. W. Jr. Rice, Map of Mars showing channels and possible paleolake basins, USGS Misc. Inv. Ser. Map 1-2461 (1:30,000,000), 1995.

Shoemaker, E. M., R. L. Squires, and Abrams, Bright Angel and Mesa Butte fault systems of northern Arizona, in Cenozoic tectonics and regional geophysics of the 172

western Cordillera, Smith R. B. and G. P. Eaton, eds.. Geologic Society of America Memoir 152, p. 341-367, 1978.

Snow, D. T., Rock fracture spacings, openings, and porosities, J. Soil Mech. Found. Div., Proc. Amer. Soc. Civil Engrs., 94, p. 73-91, 1968.

Touma, J., and J. Wisdom, The chaotic obliquity of Mars, Science, 259, p. 1,294-1,297, 1993.

U. S. Army Corps of Engineers, HEC-RAS River Analysis System, version 3.0, Hydraulic Reference Manual CPD-69, U.S. Army Corps of Engineers, Hydrologic Engineering Center, Davis, California, variable pages, 1998.

U. S. Geological Survey (USGS), Casner Butte Quadrangle 7.5 minute topographic map, AMS 3653IINW-Series V898 (1:24,000), 1965.

Webb, R. H. and R. D. Jarrett, One-Dimensional Estimation Techniques for Discharges of Paleofioods and Historical Y\oods,in Ancient Floods, Modem Hazards, Principles and Applications of Paleoflood Hydrology, House, P.K., R.H. Webb, V.R. Baker, and D.R. Levish, eds., American Geophysical Union, Water Science and Application 5, Washington, D.C., p. 111-125,2002.

Weir, G. W., G. E. Ulrich, and L. D. Nealey, Geologic map of the Sedona 30' x 60' quadrangle, Yavapai and Coconino counties, Arizona, USGS Misc. Inv. Ser. I-Map 1896 (1:100,000), 1989. 173

Figure 1. Side Looking Aperture RADAR (SLAR) image of Wet Beaver Creek and pertinent features. Location of Figure 4 is also shown, with the stream gage located approximately 2 kilometers east. 174

Figure 2. Viking image of Abus Vallis and the other sapping chaimels located in our martian study region. The blue line is an approximation of Abus Vallis' surface watershed; the thicker the line, the greater the uncertainty of its position. The green box indicates the location of the Mars Orbital Camera (MOC) imagery described in the manuscript (Image courtesy of Malin Space Sciences Systems). 175

Figure 3. Upstream view on Wet Beaver Creek. Note that the channel bends to the right, exposing the terraced sandstone that composes the channel banks.

Figure 4. Wet Beaver Creek channel floor. Terraces appear to form two distinct deposits of large cobbles and botilders, most noticeable in the background. 176

U8GS 09505350 DRV SERVER CREEK NEAR RIMROCK, ARIZ.

1S85 1978 1975 1989 1985 1998 1995 2888 DATESI 87/14/1961 to 89/23/1999

^iiSGS USG8 09505E00 UET BEfllVER CREEK NEAR RZMROCK, RRIZONfl 28888

15888

o 18888 o- <). o o o o • o o o- 5888 O ^ ^ ^ O ^ o o o o- > o o o o O o o O ^ ^ o •o- 1965 1978 1975 1988 1985 1998 1995 2988 DBTESJ 82/12/1962 to 87/27/1999

Figure 5. Peak discharges for Dry Beaver Creek and Wet Beaver Creek between 1962 and 1999, the length of the historical record. Note the larger discharges obtained by Dry Beaver Creek (Courtesy USGS). 177

Mean Monthly Streamflow 1999

O 40

g 30

1 2 3 4 5 6 7 8 9 10 11 12 Month

p»—Dry Beaver Creek Wet Beaver Creek | Figure 6. Comparison of the mean montUy streamflow for Dry Beaver Creek and Wet Beaver Creek during a typical water year 1999. Note that Dry Beaver Creek only flows during months that precipitation fell, while Wet Beaver Creek maintains a constant base flow of approximately 7.5 cfs due to groundwater feeding the system. Also note that Wet Beaver Creek displays a smaller response to equivalent precipitation events, due to its small watershed as compared to Diy Beaver Creek. 178

Figure 7. Photo of the trimline as seen from outside the inner channel (Note Casner Butte in the background). The unobscured portion of Ihe trimline has been circled by a red dashed line. It is located just north of Figure 4 (Figure 1), which is one of the few areas where it is not covered in vegetation. The trimline itself appears like a brown line from this vantage, with gray rocks and tan vegetation above and below it 179

Wet Beaver Creek Ran: Ran 04

Wet Beaver Creek Ran: Ran 04 Dovwstream ftom tetraae evidence -OS .055 4800< Legend

EGPP 1

Ground Bank Sta

4400

4200

4000

3800- 500 1000 1500 2500 3000 3500

Figure 8. HEC-RAS one-dimensional model for Wet Beaver Creek with a discharge of 1,500,000 cfs, correlating to the trimline. The cross-section shown is slightly downstream from Figure 4 (Figure 1). 180

Figure 9. Image of fissures that characterize the land surface which composes the source region of Wet Beaver Creek. Note Brunton compass for scale.

Figure 10. The quasi-Theatre head of Wet Beaver Creek. 181

APPENDIX D

DARK SLOPE STREAKS ON MARS: ARE AQUEOUS PROCESSES INVOLVED?

J. C. Ferris, J. M. Dohm, V. R. Baker, T. Maddock in

Manuscript accepted for publication in Geophysical Research Letters. 182

DARK SLOPE STREAKS ON MARS: ARE AQUEOUS PROCESSES

INVOLVED?

J. C. Ferris', J. M. Dohm^ V. R. Baker^T. Maddock III'; 'Department of Hydrology and

Water Resources, The University of Arizona, Tucson, AZ, 85721, [email protected].

ABSTRACT

Concentrations of dark slope streaks occur in the equatorial latitudes of Mars, mostly where magmatic-driven activity dominates the geologic record. Although originally ascribed to wet debris flows, all the most recent published hypotheses concerning these features focus on processes which disturb a brighter dusty mantle to expose a darker substrate. These mechanisms invoke dry mass wasting or eolian processes, excluding a role for water. In light of the geographic, geologic, and morphologic considerations, and the new information provided from the Mars Orbital Camera and the

Mars Orbital Data Altimeter, we reexamine fluvial processes as a viable explanation for some of the dark slope streaks. In our opinion, two contrasting processes for the formation of dark slope streaks, dust avalanching and spring discharge, represent endpoints on a continuimi of progenitors. It may be that some of these features result from dry mass wasting or eolian processes, some from fluvial processes, and some from a mechanism(s) not yet conceived. A spring discharge origin for the formation of the dark slope streaks has profound implications, including Mars having limited, but currently active, fluvial processes acting upon its surface, as well as near-surface aquifers. 183

INTRODUCTION

Mars Orbital Camera (MOC) images portray concentrations of dark slope streaks

non-uniformly distributed within the equatorial region (±30° latitude zone) of Mars

[Riflcen and Mustard, 2001]. Since the Viking missions of the 1970s, several hypotheses

for the formation of these features have been presented, including debris weathered j&om

dark block inclusions within pyroclastic materials [Morris, 1982], stains from wet debris

flows that may involve brines [Ferguson and Lucchitta, 1984], the disturbance of a dust mantle by mobilized imderlying debris that results in darker, dust-deficient scars [Williams,

1991], and dust avalanching resulting from the oversteepening of airfall deposits [Sullivan et al., 2001]. hi the last fifteen years, all of the published hypotheses concerning these features have focused on their being solely the result of eolian or dry mass-wasting processes. We wish to reevalmte the water-related hypothesis, first posed by Ferguson and

Lucchitta [1984], because current models do not seem to explain the total spectrum of dark slope streak morphologies observed on Mars.

GEOGRAPfflC AND GEOLOGIC SETTING OF THE DARK SLOPE STREAKS

The largest anomalous concentration of dark slope streaks occurs west of the

^gantic northeast-trending chain of shield volcanoes of Tharsis Montes, in the Mangala

Valles [Scott and Tanaka, 1986] and the northwestern slope valleys (NSVs) [Dohm et al.,

2001] regions, along the margiii of the highland-lowland boimdaiy (Figure 1). An abrupt change in slope occurs to the east and northeast of the Mangala Valles and the NSVs regions as a result of the drainage-divide-fonning shield volcanoes and associated lava 184 flows of Tharsis Montes and Olympus Mons, respectively (Figure 1). Generally located throughout the southern part of the region that contains the greatest concentration of dark slope streaks. Medusae Fossae Formation materials form a discontinuous band of wind-etched materials along the highland-lowland boundary [e.g., Scott and Tanaka, 1986].

On the basis of its morphology, the Medusae Fossae Formation has been interpreted to consist of ash-flow tuffs [e.g., Malin, 1979] among several potential origins [e.g., Schultz and Lutz, 1988].

Intriguingly, other concentrations of dark slope streaks occur only within or near regions with geologic histories marked by Late Hesperian and yoimger magmatic, tectonic, and fluvial activity, generally within the ±30° latitude zone [Rifkin and Mustard, 2001]

(Figure 1), including the Tharsis Montes shield volcanoes [e.g., Scott and Zimbelman,

1995] and their margins [Scott and Tanaka, 1986], Arabia Terra located west of the lava-forming rise of Syrtis Major [Greeley and Guest, 1987], and the Elysium rise comprised of several large shield volcanoes and fissure fed eruptions [Mouginis-Mark,

1985]. All the above regions have been proposed to contain geomorphic evidence that indicate potential ground ice/magma interactions.

RATIONALE FOR THE RECONSIDERATION OF FLUVIAL PROCESSES

Although we agree that a portion of the dark slope streak features most likely result fiom dry mass wasting or eolian processes, in our opinion, some of their characteristics are not best explained by dry mass wasting or eolian processes (Table 1). For example, if all of these features are the result of disturbing a brighter dusty mantle to expose a darker 185 substrate, why are they generally not uniformly present in regions with similar materials, topography, gradients, and low thermal inertia? Significantly, there are no observable dark slope streaks located in older Noachian and Early Hesperian terrains of similar geology, topography, and low thermal inertia, examples include: (1) the summit and flanks of

Apollinaris Patera, a shield volcano that developed on volatile-rich substrates [Scott et al.,

1993], (2) the networking troughs that dissect Late Noachian and Early Hesperian ftiable material, interpreted to be ignimbrites, near the southeastern margin of the Thaimiasia plateau [Dohm and Tanaka, 1999], and (3) parts of the polar regions and Tharsis. One explanation posed for the absence of these features in regions of similar materials, topography, and low thermal inertia is that dust does not fall out at the same rate everywhere [Sullivan et al., 2001]. However, the correlation of the dark slope streaks to specific geologic environments and histories as portrayed on published geologic maps and by published paleohydrologic research bears significantly on the interpretation of the origin for some of these features.

The albedo variation among dark slope streaks (Figures 2 - 4) is explained by a gradual obscuration by a mantling of dust [Williams, 1991]. As the features become mantled, they appear lighter in contrast to their surroimdings. In our opinion, this mechanism doesn't account for the uniform albedo of each feature, especially where the extremely long features are concemed. With the mantling of dust, we would expect to see such features with a varying thickness of mantling material along their length. Due to chaotic wind patterns resulting fi-om various factors, including local topographic variation, some areas would receive more dust than others. Little is known about where and at what 186 rate dust is falling out on Mars, but given the complexity seen in atmospheric and geologic processes studied on Earth, it seems unlikely that all streaks should have a constant albedo throughout their length.

In light of the above considerations, we propose that fluvial processes should be reconsidered as a viable explanation for some of the dark slope streaks. In our opinion, dust avalanching and spring discharge represent endpoints on a continuum of progenitors.

It may be that some of these features result from diy mass wasting or eolian processes, some from fluvial processes, and some from a mechanism(s) we have not yet conceived.

THE SPRING DISCHARGE HYPOTHESIS

We propose that the dark slope streaks result from ephemeral groimdwater springs derived from ground ice, a hypothesis that significantly modifies the hypothesis originally presented by Ferguson and Lucchitta [1984]. Groimdwater seepage from springs flows down the slope faces, where an appreciable amount of this surface flow may infiltrate into the subsurface, depending upon the lithologic properties of the surface material (e.g., infiltration rate), and move as throughflow. As the spring ceases to issue, flie remaining overland flow either evaporates/sublimates or infiltrates into the subsurface. Depending on chemical composition, tiie infiltrating water may saturate the surface as interstitial water or ice. The ^ring flow may wash away a lighter dust layer and expose a darker substrate, or it may saturate the surface materials, causing the approximately 10% albedo difference.

If the saturation by interstitial water and/or ice is causing the darkened albedo, the ev^}oration and/or sublimation of that same water and/or ice would explain the gradual 187

and uniformly distributed lightening that occurs. This model also allows for the instances

where these features achieve great lengths (e.g., 1500 meters) without having to invoke

unusual transport mechanisms, such as local collapse of the martian regolith and the

release of CO2 which subsequently mixes with dust particles to produce a "dusty" density

flow [Albin and King, 2001].

Our hypothesis genetically links depressions, often forming pit crater chains near

some of the source areas of dark slope streaks (Figure 4), to the sublimation of volatiles and

the erosive action of fluid flow. It also explains features with upslope ends that are not

acute, but stretch broadly, which is consistent with water being forced through a permeable

contact between two geologic layers and/or structure (e.g. faults, fiactures, and joints).

Most importantly, in our opinion, is that a spring discharge explanation most aptly explains

cases where a dense concentration of features occurs on only one side of a valley. The

features result from springs caused by the local groundwater flow emerging from the valley

face. We will now discuss the two most plausible mechanisms for groundwater springs to exist on the surface of Mars.

Hydrothermal Systems. The potential existence of hydrothermal activity on Mars

has been suggested based on geologic investigations [Newsom, 1980; Dohm and Tanaka,

1999]. A large intrusion could provide tiie heat to drive a regional hydrotbermal system

[Gulick, 1993] in the areas where the dark slope streaks are found (Figure 1). These magmatically-diiven groundwater systems would presumably be actively circulating due to temperature, pressure, and elevation gradients. Furthermore, a near-surface interstitial ground-ice confining layer caused by surface conditions would allow for artesian 1S8 conditions to arise. The slope faces of the valleys, cliffs, and craters must either exist beneath the surface of the confining layer or, more likely, represent areas where artesian pressure exploits faults, joints, or fractures. Once a spring occurs, its ephemeral nature would most likely be the result of the spring sealing itself due to a drop in pressure and/or freezing due to atmospheric conditions. A simple explanation for why we don't see these features in regions where such potential hydrothermal activity may have existed along with similar topographies, slopes, and thermal inertia values, is time. Regions such as Warrego

Valles (including on the slopes of the valley tributaries) and the slopes of the networking trough systems of the southem Coprates rise, for example, record ancient (Late

Noachian/Early Hesperian) activity [Dohm and Tanaka, 1999] and has either imdergone a thickening of the cryosphere due to the lengthy absence of magmatic-driven fluvial activity or that the same activity may have resulted in a devolatilization of the region.

Hypersaline Aquifers. Terrestrial aquifer systems usually contain impxnre H2O with trace amounts of various salts and minerals leached from surrounding geologic materials. This is especially true for arid regions, where the slow circulation of groimdwater results in mineralization. Knauth et al. [2001] proposed that water could exist in liquid form at the martian surface under current temperature and pressures if the water was sufficiently rich in dissolved salts. The concept of a hypersaline near surface aquifer that intox^epts slope faces ^?ees with observed characteristics of the dark slope streaks. However, the question arises as to the flow behavior of hypersaline groundwater, given its increased spedflc gravity due to its mineral content Using the properties of the DEVIL brines on Mars

[Knauth et al., 2001], we sought to compare flow behavior to potable groimdwater on Earth. 189

Assuming one-dimensional flow, and an isotropic and homogeneous medium, we sought to compare the hydraulic conductivities (coefficients of permeability) of DEVIL brines on

Mars and potable water on Earth. The hydraulic conductivity is a function of the porous medium and the fluid, and is equivalent to

(k *P *g)/^ where k is the intrinsic permeability, p is the mass density, g is gravity, and jx is the dynamic viscosity. The mass density of DEVIL brines is approximately 1.3 to 1.4 g/cm^

[Paul Knauth, personal cortmunicatior{\. The dynamic viscosity of DEVIL brines is unknown, and increases dramatically as temperature decreases. Very little research has been done on the viscosity of brines at low temperatures. Paul Knauth and his colleagues

{personal commmication] noted they were gel-like when they tried to freeze them in liquid nitrogen cold traps (-195 degrees Celsius). In the absence of any data on this matter, we will have no choice but to use to dynamic viscosity of water for both the comparative equations, which is 0.011404 g/(s » cm^).

For our comparison, we will use an equivalent medium, consolidated rock, with an intrinsic permeability of 2.66 *10"'^ cm^. The mass density of terrestrial water is 0.999099 g/cm^. The hydraulic conductivity for potable water on Earth under these conditions is

2.28 * 10"^ cm/s. The hydraulic conductivity for DEVIL brines on Mars under these conditions is 1J25 * 10"® cm/s. The difference is ahnost negligible, but the viscosity was kept equivalent Once a realistic value for the dynamic viscosity of DEVIL brines is detennined, more accurate values of hydraiilic conductivity can be determined, and given observations that the DEVIL brines became gel-like at -195 degrees Celsius; the 190 differential between the two hydraulic conductivities will probably become amplified.

However, given the range of surface and near-surface martian temperatures, which are

much warmer than -195 degrees Celsius, we hypothesize the xiltimate differential between

hydraulic conductivities will be inconsequential.

The absence of perceptible erosion in these features might be explained by a spring

issuing hypersaline groundwater, whose higher mass density and viscosity would result in a lower velocity and perhaps less erosion as it moved slowly downslope. The presence of dark slope streaks in equatorial regions argues that ambient surface conditions may also play a role in allowing near-surface water to exist, perhaps only seasonally. Most noteworthy is that hypersaline or extensively mineralised aquifers often result from hydrothennal systems; both could be a factor for any given spring.

IMPLICATIONS

The implications of the dark slope streaks being the result of spring discharge processes are not without significance. Given how quickly these features have been witnessed to come into existence [Edgett et al., 2000], it would mean that Mars has limited, but currently active, hydrological processes acting on its surface. Furthermore, features that can be remotely verified by the spectral signature of water vapor would mark ideal locations for sample retum missions, manned exploration, and the study of fossil or extant life. 191

ACKNOWLEDGEMENTS

We thank Rob Sullivan for his stimulating discussion and incisive criticism (not all of which led to ultimate agreement). This research was supported in part by NASA Grant

NAG-5-9790. 192

REFERENCES

Albin, E. F., and J. D. King, Dark slope streaks and associated layered deposits on the southwestern floor of Cassini impact basin. Mars, Lunar Planet. Sci. Conf.[CD-ROM], XXXn, abstract 1380,2001.

Dohm, J. M., and K. L. Tanaka, Geology of the Thaumasia region. Mars: plateau development, valley origins, and magmatic evolution. Planetary and Space Science, 47, 411-431,1999.

Dohm, J. M., R. C. Anderson, V. R. Baker, J. C. Ferris, L. P. Rudd, T. M. Hare, J. W. Jr. Rice, R.R. Casavant, R. G. Strom, J.R., Zimbelman, and D. H. Scott, in press. Latent activity for western Tharsis, Mars: significant flood record exposed, JGR Planets, 106, 12301-12314,2001.

Edgett, K. S., M. C. Malin, R. J. Sullivan, P. Thomas, and J. Veverka, Dynamic Mars: new dark slope streaks observed on annual and decadal time scales. Lunar Planet. Sci. Conf. [CD-ROM], XXXI, abstract 1058,2000.

Ferguson, H. M., and B. K. Lucchitta, Dark streaks on talus slopes. Mars, Reports of the Planetary Geology Program 1983, NASA TM-86246,188-190,1984.

Greeley, R, and J. E. Guest, Geologic map of the eastern equatorial region of Mars, USGS Misc. Inv. Ser. MqpI-1802B (L-15.000,000), 1987.

Gulick V. C., Magmatic intrusions and hydrothermal systems: Implications for the formation of small Martian valleys, Ph.D. Utesis, Univ. of Arizona, Tucson, Ari2»na, 1993.

Knauth, L., P., D. M. Burt, and S. Klonowski, Dense, eutectic, valley-forming, intermediate latitude (DEVIL) brines on Mars, Geol. Soc. America Abstracts with Programs, v. 32, no. 7, p. A303,2000.

Malin, M. C., Mars: evidence of indurated deposits of fine materials, (abstract), NASA Conf Pub. 2072,54,1979. 193

Morris, E., Aureole deposits of the Martian volcano Olympus Mons, J. Geophys. Res., 87, 1164-1178, 1982.

Mouginis-Mark, P. J., Volcano/ground ice interactions in Elysium Planitia, Mars, Icarus, 64,265-284,1985.

Newsom, H. E., Hydrothermal alteration of impact melt sheets with implications for Mars: Icarus, 44,207-216,1980.

Rifkin, M. K., and J. F. Mustard, Global distribution of unique surface processes imaged by the Mars Orbital Camera, Lunar Planet. Sci. Conf. [CD-ROM], XXXn, abstract 1698, 2001.

Schultz, P. H., and A. B. Lutz, Polar wandering on Mars, Icarus, 73,91-141,1988.

Scott, D. H., and K. L. Tanaka, Geologic map of the western equatorial region of Mars, USGS Misc. Inv. Ser. MapI-1802-A (1:15,000,000), 1986.

Scott, D.H. and J. R. Zimbehnan, Geologic map of Arsia Mons volcano. Mars, USGS Misc. Inv. Ser. Map 1-2480 (1:1,000,000), 1995.

Scott, D. H., J. M. Dohm, and D. J. Applebee, Geologic map of science study area 8, Apollinaris Pataa region of Mars, USGS Misc. Inv. Ser. Map 1-2351 (1:500,000), 1993.

Sullivan, R., M. Maiin, and K. Edgett, Mass-Movement Slope Streaks Imaged by the Mars Orbital Camera, JGR Planets, 106,23607-23633,2001.

Williams, S. H., Dark talus streaks on Mars are similar to aeolian dark streaks. Lunar Planet. Sci. XM, 1509-1510,1991. Elysium: ehrySe

iPlanftia • r"Monsi^ v

oUlnarfe tera T.:a: wV*!

Figure 1. MOLA silacled relief map (MOLA Science Team). General locations of anomalous concentrations of dark slope streaks indicated by hatchured pattern (adapted from Rifkin and Mustard, 2001). Warrego Valles and southern Coprates rise regions are indicated by W.V. and s.C.r., respectively.

Figure 2. MOC image M09-03285. Shown are the continuation of dark slope streaks onto valley floors, as well as their varying shapes and albedos. Figure 3. I\/10C image M07-04393. Dark slope streaks within an elongated depression, which source at varying gradients (e.g., A and B) and elevations. Note anastomosing patterns (e.g., C) and digitate termini.

Figure 4. MOC image M07-04393. Dark slope streaks on the slope of a tributary to Mangala Valies. Note the dark slope streak that abutts the rim of a small impact crater (e.g., D). This is interpreted to be infiltration into the fractured rim materials, perhaps as throughflow. Also note the cavi (depressions) on the top of the tear-drop shaped mesa (e.g., E). These are interpreted to be the result of volatile releases. 196

Proposed Dark Slope Streak Models

Dry Mass Spring Dark Slope Streak Wasting or Discharge Combination of Characteristic Eolian orocesses processes both processes

Predicted to have many instances where dark slope streaks occur on only one side of a valley? No Yes Yes Predicted to source at geologic contacts? No Yes Yes

Predicted to source near structure (e.g. faults, fractures, and joints')? No Yes Yes Predicted to leave no depositional evidence behind? Maybe Maybe Maybe Typically between 300 and 1500 meters in length? Yes Yes Yes Predicted to occationally be much greater in length (—3 km)? Maybe Yes Yes Triangular, apex-up shapes? Yes Yes Yes Constant albedo within streak? Maybe Yes Yes Constant albedo within long streaks (e texture? Maybe Maybe Maybe Consistant with recent activity in current surface environment? Yes Unknown Yes Predicted to occur over a wide variety of geologic imits? Yes Maybe Yes Predicted to occur where magmatic driven activity and paleohydrologic activity dominates the geologic record? No Yes Yes Predicted to be associated with cavi and depressions? No Yes Yes Predicted to only occur in areas of low thermal inertia? Yes No Yes Maybe, if Predicted to occur across a wide hydraulic head is altitude range, including sufScient or elevations of 15 km? Yes aquifer is perched Yes Associated with areas of great layered ^ecta crater density? No Yes Yes

Table 1. Comparison of multiple formation processes. 197

CURRICULUM VITAE JUSTIN C. FERRIS

8701-21 E. Tanque Verde Road Tucson, AZ 85749 Work Tel: (520) 370-6357 Home Tel; (520) 887-3365 E-mail: [email protected]

Education

Ph.D., Department of Hydrology and Water Resources, The University of Arizona 2002 Major Hydrology Minor: Geology Dissertation: Treatises on the Paleohvdrologv of Mars and Earth Principle Advison Professor Victor R. Baker

B.S., Dq)artment of Geology, Northern Arizona University 1997 Major: Geology Emphasis: Hydrogeology

Research Experience

Research Associate 1999-2002 Department of Hydrology and Water Resources, The University of Arizona, Tucson, Arizona I was the resident hydrologist/geomorphologist for an interdisciplinary group of investigators conducting research on the geologic, hydrologic, and climatological history of Mars. My responsibilities included, but were not limited to, the interpretation of geomorphic features based on remote sensing and non-traditionai data sets. Once I determined the features of interest to be hydrogeomoiphic in origin, I conducted paleohydrologic and/or hydrologic studies in coordination with our comprehensive research goals. In the course of this research, I had to employ numerous imiovative methodologies to redejBne hydraulic flow equations to account for martian gravity and pressures, and to design my own hydrologic models accordingly. These efforts resulted in numerous publications in peer-reviewed journals, as well as significant media attention regarding our findings.

Hydrogeomorpliologist 20G1 JE Fuller/ Hydrology & Geomorphology, Inc., Tempe, Arizona I worked on various hydrologic and geomorphic projects, including the Agua Fria and Skunk Creek Watershed Management Plans, the Carefi^ Drainage Master Plan, and various projects for the Flood Control District of Maricopa County. Examples of 198

work done includes using GIS (i.e., AutoCAD and RiverCAD) to determine the historical behavior of jSve tributaries to the Cave Creek drainage, and field work to characterize the fluvial geomorphology of these tributaries.

Research Assistant 1998 Department of Geosciences, The University of Arizona, Tucson, Arizona Sole operator of laboratory where I extracted benzene from samples for use in radiocarbon dating.

Intern Hydrologist 1997 Arizona Department of Water Resources, Hydrology Division, Phoenix, Arizona Conducted field research in both surface and subsurface hydrology, as well as water table impact analysis modeling. Created instream flow and augmentation grant databases.

Teaching Experience

Teaching Associate 2000-2902 Department of Planetary Sciences, The University of Arizona, Tucson, Arizona Sole teaching associate for Universe and Humanity, Ori^n and Future (instructor: Tom Gehrels, ecoroUment: 60) Spring 2001 and 2002. Sole teaching associate and lab instructor for Geology of the Solar System (instructor Robert Strom, enrollment: 20) Fall 2001. Department of Tree-Ring Research, The University of Arizona, Tucson, Arizona Teaching associate for Introduction to Global Change (instractor: Katherine Hirschboeck, enrollment: 130) Spring and Fall 2000.

Teaching Assistant 1998-1999 Department of Hydrology of Water Resources, The University of Arizona, Tucson, Arizona Teaching assistant and sole lab instructor for Water and the Environment (instructors: Robert Clark and Robert MacNish, enrollment: 130) Fall 1998 and Spring 1999.

Tutor 1997-1998 S A.L.T. Center, The University of Arizona, Tucson, Arizona Tutored learning disabled students in hydrology, geology, oceanography, soil & water science, and mathematics. 199

Other Experience

Summer Orientation Faculty Advisor 2000-2001 University College, The University of Arizona, Tucson, Arizona Advised incoming freshmen and non-traditional students in career planning and course selection. Summer 2000 and 2001.

Honors and Awards

1999-2002 Computer Science, Engineering, and Mathematics Scholarship; National Science Foimdation 1999 Summer Field Camp Scholarship, Kenneth D. Schmidt and Associates

Professional Memberships

American Geophysical Union Arizona Hydrological Society Geological Society of America National Ground Water Association

Peer Reviewed Scientific Publications

Ferris, J.C., J. M. Dohm, V. R. Baker, and T. Maddock in. Dark Slope Streaks on Mars: Are Aqueous Processes Involved?, accepted for publication in Geopf^sical Research Letters, 2002.

Dohm, J. M., J. C. Ferris, V. R. Baker, R. C. Anderson, T. M. Hare, R, G. Strom, N. G. Barlow, K. L. Tanaka, J. E. Klemaszewski, and D. H. Scott, Ancient Drainage Basin of the Tharsis region. Mars: Potential source for outflow channel systems and putative oceans or paleolakes. Journal of Geophysical Research, Vol. 106, p. 32,943-32,958,2001.

Dohm, J. M., R- C. Anderson, V. R. Baker, J. C. Ferris, L. P. Rudd, T. M. Hare, J. W. Rice, Jr., R. R- Casavant, R. G. Strom, J. R_ Zimbelman, and D. H. Scott, Latent outflow activity for western Tharsis, Mars: Significant flood record exposed. Journal of Geophysical Research, Vol. 106,12,301 —12,314,2001.

Dohm, J. M., R. C. Anderson, V. R. Baker, J. C. Ferris, T. M. Hare, R. G. Strom, L. P. Rudd, J. W. Rice, Jr., R. R. Casavant, and D. H. Scott, System of gigantic valleys northwest of Tharsis, Mars: latent catastrophic flooding, northwest watershed, and implications for northern plains ocean. Geophysical Research Letters, Vol. 27, No. 21, 3,559 - 3,562,2000. 200

Scientific Publications in Review

Ferris, J. C., J. M. Dohm, and T. M. Hare, Comparative Hydrogeomorphology of Wet Beaver Creek, Arizona, and Abus Vallis, Mars, submitted to Earth Surface Processes and Landforms, 2002.

Conference Presentations and Publications

"The Vast Continuum of Martian Hydrology: Ancient Catastrophic Floods and Present- Day Hydrogeomorphic Processes." Invited talk for the Geology Department's Spring Seminar Series, Northern Arizona University, Flagstaff, Arizona, 2002.

"Daric Slope Streaks on Mars: Are Aqueous Processes hivolved?" Paper presented at the Annual Symposiimi of the Arizona Hydrological Society, Tucson, Arizona, 2001.

"Dark Sloping Features: Evidence of Current Fluvial Activity on Mars?" Paper presented at the 11*'' Annual El Dia del Agua Symposium on Hydrology, Tucson, Arizona, 2001.

Selected Abstracts

Dohm, J. M., S. Maruyama, V. ?_ Baker, R. C. Anderson, J. C. Ferris, and T. M. Hare, Plate Tectonism on Early Mars: Diverse Geological and Geophysical Evidence, The Lunar and Planetary Science Conference XXXHL [CD-ROM abstract 1639], Houston, Texas, March 11-15,2001.

Dohm, J. M, S. Maruyama, V. R. Baker, R. C. Anderson, J. C. Ferris, and T. M. Hare, Evolution and Traits of Tharsis Supeiplume, Mars, International Workshop: Role of Superplume in Earth System from central core to the surface including evolving Life Abstracts, Tokyo, Japan, January 25 - 28,2002.

Dohm, J. M, S. Maruyama, V. R. Baker, R. C. Anderson, and J. C. Ferris, Evolution of the Tharsis Magmatic Complex: Traits of a Terrestrial Superplume (2), GSA Annual Meeting Abstracts, Boston, Massachusetts, November 5 - 8,2001

Dohm, J. M., J. C. Ferris, V. R. Baker, R. G. Strom, R. C. Anderson, T. M. Hare, K. L. Tanaka, J. A. Skinner, N. G. Barlow, and J. E. Klemaszewski, 3-D Animation of the Evolution of the Tharsis Magmatic Complex, Mars, GSA Cordilleran Section Abstracts, Universal City, California, April 9-11, 2001. 201

Anderson, R. C., J. M. Dohm, M. Golombek, V. R. Baker, J. C. Ferris, and T. M. Hare, Amazonian Faulting: Is Mars Tectonically Active Today?, The Ltmar and Planetary Science Cortference XXXII [CD-ROM abstract 2130], Houston, Texas, March 12 -16,2001.

Baker V. R., R- G. Strom, J. S. Kjirgel, J. M. Dohm, and J. C. Ferris, Very Recent, Water- Related Landforms on Mars, The Lunar and Planetary Science Conference XXXII [CD-ROM abstract 1619], Homton, Texas, March 12 -16,2001.

Dohm, J. M., J. C. Ferris, R. C. Anderson, V. R Baker, T. M. Hare, N. G. Barlow, R. G. Strom, K. L. Tanaka, and D. H. Scott, Paleotopographic Reconstruction of the Tharsis Magmatic Complex Reveals Potential Ancient Drainage Basin/ Aquifer System, The Lunar and Planetary Science Cortference XXXII [CD-ROM abstract 1870], Houston, Texas, March 722001.

Dohm, J. M., J. C. Ferris, V. R. Baker, R. G. Strom, R. C. Anderson, T. M. Hare, K. L. Tanaka, J. A. Skinner, N. G. Barlow, and J. E. Klemaszewski, Paleotopographic Reconstruction of the Principle Magmatic Complex of Mars, Tharsis, Exposes Gigantic Drainage Basin, GSA Annual Meeting Abstracts, Reno, Nevada, November 9 -18,2000.

Dohm, J. M., J, C. Ferris, RC. Anderson, V. R. Baker, T. M, Hare, K. L, Tanaka, N. G. Barlow, D. H. Scott, J. A. Skinner, and J. E. Klemaszewski, The Tharsis Basin of Mars: A Potential Source for the Outflow Channel Systems, Layered Wall Deposits of Valles Marineris, and Putative Oceans or Paleolakes, Tharsis Workshop Abstracts, Keystone, Colorado, October 16-18, 2000.

Dohm, J. M., R-C. Anderson, V. R. Baker, C. R, Neai, J. C. Ferris, T. M. Hare, K. L. Tanaka, D. H. Scott, J. A. Skinner, and J. E. Klemaszewski, Evolution of the Principal Magmatic Complex of Mars, Tharsis, Tharsis Workshop Abstracts, Keystone, Colorado, October 16-18, 2000.

Dohm, J. M., V. R Baker, RC. Anderson, V. R. Baker, J. C. Ferris, T. M. Hare, K. L. Tanaka, J. E. Klemaszewski, D. H. Scott, and J. A. Skinner, Martian Magmatic-driven Hydrothermal Sites: Potential Sources of Energy, Water, and Life, Tharsis Workshop Abstracts, Keystone, CQlorado, October 16 -18. 2000.

Tanaka, K. L., J. M. Dohm, J. C. Ferris, and R. C. Anderson, Erosional Mechanism for the Development of Valles Marineris, Tharsis Workshop Abstracts, Keystone, Colorado, October 75-75,2000.

Dohm, J. M., R- C. Anderson, V. R. Baker, J. C. Ferris, T. M. Hare, R. G. Strom, L. P. Rudd, J. W. Rice, Jr., and D. H. Scott, Northwestern Tharsis Latent Outflow Activity on Mars, The Second International Conference on Mars Polar Science and Exploration, Reylgavik, Iceland, August 21 —25, 2000.