University of Southern California Dissertations and Theses

Late Quaternary erosional and depositional history of Sierra del Mayor, Baja California, Mexico

(USC Thesis Other)

Late Quaternary erosional and depositional history of Sierra del Mayor, Baja California, Mexico

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content LATE QUATERNARY EROSIONAL AND DEPOSITIONAL HISTORY OF SIERRA DEL MAYOR, BAJA CALIFORNIA, MEXICO by Louis David Carter A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Geological Sciences) February 1977 UMI Number: DP28541 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. Dissertation Publishing UMI DP28541 Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code u e s t* ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106- 1346 UNIVERSITY OF SOUTHERN CALIFORNIA T H E G R A D U A TE S C H O O L U N IV E R S IT Y PA R K LOS A N G E LE S . C A L IF O R N IA 9 0 0 0 7 This dissertation, w ritte n by Louisi D a v id _ C a rt_ e r under the direction of A i . ? . . . . Dissertation Com m ittee} and approved by a ll its members} has been presented to and accepted by The Graduate School, in p a rtia l fu lfillm e n t o f requirements of the degree of D O C T O R O F P H IL O S O P H Y Dean _ J a n u a ry 1 3 , 1977 D ate..................... DISSER ATIO N CO M M ITTEE Chairman ■Ph.D. Gre. ' 7 7 C 3 2H- J t & o o c , ZV\, CONTENTS Page ABSTRACT.............................................. xii INTRODUCTION ......................................... 1 General ......................................... 1 Description of the problems .................. 6 Regional significance......... ............... 16 Scope and methods • • • • • • • • • • • • • • 18 Previous w o r k .................... 18 Acknowledgments • • • • • « • • • • • • • • • 19 DESCRIPTION OF THE A R E A ............................ 21 Physiographic setting ......................... 21 Geologic setting ................................ 25 Stratigraphy and lithology ................ 25 Structure and tectonic development • • « • 31 Cenozoic history of the Colorado River D e l t a ..................................... 3 4 Climate......................................... 33 Flora'............................................ ^0 Culture......................................... 4l Boundaries and accessibility ••••••••« h2 ii Page GEOLOGY.............................................. kk Introduction . . . . . . . . . . . . . . . . hh Geomorphology • • • • • • • • • • • • • • • • hh Mountain block • • • • • • • • • • • • • h'J Piedmont slopes • • • • • • • • • • • • • 5^ Basin floors • • • • • • • • • • • • • • 67 Stratigraphy • • • • • • • • • • • • • • • • 67 Basement r o c k s .................... 68 Metamorphic rocks •••••••••• 68 Granitic rocks ................ 69 Older alluvium • • • • • « • « • » • • • 70 Red Ridge section •••••••••« 71 Older piedmont gravels .•••••• 75 Younger alluvium « • • • • • « • • • • • 79 Structure • • • • • • • • • • • • • • • • • • 80 Mountain block • • • • • • • • • • • • • 80 Range front faults ••••••••••• 80 Piedmont faults • • • • • • • • • • • • • 82 FINE-GRAINED PIEDMONT SEDIMENTS .................. 88 Distribution • • • • • • • • • • • • • • • • 88 Derivation • • • • • • « • • • • • • • • • • 88 Description « • • • • » • • • • • • • • • • • 91 Correlation, age and implications . . . . . . 105 Page VALLEY FILL......................................... 112 Introduction • • • • • • • • • • • • • • • • 112 Episodes of filling........................... 113 Depositional processes ••••••••••• 117 Source of debris and nature of debris production • • • • • • • • • • • • • • • • 122 Timing and cause of mountain block alluviation • •••••••••• 123 RELATION OF THE UPPER SURFACE OF VALLEY FILL AND SMOOTH-SURFACED PIEDMONT TERRACES .............. 128 Introduction • • • • • • • • • • • • • • • • 128 Relation of the upper surface of valley fill to smooth-surfaced piedmont terraces developed on alluvium ....................... 130 Eastern and southeastern piedmont and tributary basins ••••••••••• 130 Southern piedmont and tributary basins . 130 Valle Abanico and Valle Abanico fan . 130 Relation of the upper surface of valley fill to smooth-surfaced pediment terraces • • • 137 Rock Fan and Barren V a l l e y .............. 137 Pediment Canyon ........................... 1^3 Red Ridge and Canon de Culebras • • • • • 1^+7 Terrace ! • • • • • • • • • • • • • • 150 iv Page Terrace XI • • • • • • • • • • • • • • 156 Terrace I I I ............................ 156 Canon de Culebras •••••••••• 163 Summary and discussion • • • • • • • • • • • • 171 AGE, ORIGIN AND IMPLICATIONS OF THE DESERT PAVEMENT SURFACES ................................ 175 Relative a g e ..................................... 175 Origin • • • • • • • • • • • • • • • • • • • • 177 Implications • • • • • • • • • • • • • • • • • 179 MOUNTAIN BLOCK INCISION AND THE REMOVAL OF VALLEY F I L L ....................................... I83 Introduction ..................................... 183 General description • • • • • • • • « • • • • IS3 Cause • • • • • • • • • • • • • • • • • • • • 184 MORPHOLOGIC DEVELOPMENT OF THE PIEDMONT ......... 190 Introduction ..................................... 190 Southern piedmont .............................. 190 Introduction ................................ 190 Upper piedmont s l o p e s ..................... 190 Distribution of surface types • • • • 191 Topographic profiles ••••••••• 196 Origin of segmentation ................ 205 Discussion.............................. 210 Fan evolution......................... 219 Lower piedmont slopes •••••••••• 223 v Page Western piedmont •«««•««« .............. 229 Eastern piedmont ......................... 231 ORIGIN OF THE TERMINUS OF SMOOTH-SURFACED TERRACES............................................ 232 Introduction . # ................................ 232 South-eastern terminus............. 232 Southern terminus .............................. 235 Western terminus ••••«••••« ......... 237 Eastern terminus.................... 238 An original depositional feature ......... 238 Marine erosion .............................. 239 Faulting........... 240 Erosion by the Colorado R i v e r ........... 242 SUMMARY OF LATE QUATERNARY HISTORY AND A CLIMATIC INTERPRETATION ..................................... 245 Summary......................................... 245 Climatic interpretation 249 REFERENCES CITED ..................................... 254 ILLUSTRATIONS Figure Page 1, Location of Sierra del Mayor and boundary of the study area.............................. 2 2, Physiographic features in the vicinity of the study a r e a ............................ . • 4 3* The single terrace along the east side of Sierra del Mayor .............................. 7 4, Terrace fragments of the southern piedmont . 9 3* Multiple terraces of the western piedmont . . 9 6. Relict valley floor in Valle de Leneros . . . 12 7* Piedmont terrace at Sierra de las Pintas • • l4 8, Boundaries of the Colorado Desert............ 22 9* Composite stratigraphic section of the Imperial Valley region ....................... 27 10. Geomorphic subdivisions of the study area • • 45 11. Drainage basins and major drainage of the mountain block ..... 48 12. Microrelief on the upper level of valley fill in Valle Abanico......................... 52 13* Relict bar and channel morphology on benches below the upper level of valley fill in Camp Canyon..................................... 52 14. Escarpment in alluvium at the distal edge of terrace remnants ............................ 56 15. The portion of the southern piedmont with fan morphology................................ 58 vii Figure Page 16. Lower slopes of the southern piedmont • • . • 58 17* Aerial view of the western piedmont.......... 63 18, Fault scarp in alluvium in the western piedmont • • • • • • • • • * , • • • • • • • 63 19* Terrace surfaces and remnants of older materials between Red Ridge and the range front............................................ 65 20, Red Ridge • • • • • • • • • • • • • * • • • • 65 21, Tilted beds of the Red Ridge section and the eastward sloping surface that caps them • • • 72 22, Inferred age of last movement on piedmont faults •••,••••••, •«««•«.. 84 23• Fans and fan forming washes of the southern range m a r g i n .................................. 86 24, Areas of exposed older Colorado River alluvium and the locations of samples and sections discussed in the text ••,,,,, 89 25• Characteristics of local alluvium and flood plain deposits................................ 92 26, Diagrammatic sketch of facies relations at the southeastern edge of the piedmont • • • • 96 27• Margin of a broad filled channel ,••••• 98 28, Facies relations at the southwestern corner of the piedmont • • • • • • • • • • • • • • • 101 29, Direction of dip of ripple cross lamination , 103 30, Flood plain sedimentation of a former Colorado River delta 106 31, Lateral and vertical homogeneity of valley f i l l ............................................ 114 32, Diagrammatic sketch of the occurrence of stratification in valley fill................ 118 viii Figure Page 33. Chaotic deposits at the head of Camp Canyon 120 34. Crudely stratified deposits in Canon de Culebras....................................... 120 35• Relation between valley fill and flood plain deposits at Valle de Leneros ..... 124 36. Location of topographic profiles and areas of the southern piedmont mentioned in the text............................................ 131 37* Correlation of the highest valley terrace and highest fan terrace, Valle Abanico and Valle Abanico f a n ............................ 134 38. Continuity of the upper level of valley fill in Barren Valley with a pediment at the valley mouth .................................. 139 39. Rock fan; a fan-shaped rock-cut surface • • l4l 40. Pediments at Pediment Canyon................. 14.5 41. Terraces of the western piedmont and the location of topographic profiles ........... 148 42. Terrace I, Red Ridge area, profile C-C1 . . 152 43. Terrace I, Red Ridge area, profile D-D* . . 154 44. Terrace XX, Red Ridge area................... 157 45. Terrace III, Red Ridge a r e a ................ 159 46. Lag-gravel of Terrace III overlying strati fied gravels near Red R i d g e ................ l6l 47* The Terrace III surface (middleground) . . . 164 48. Terraces in Canon de Culebras.............. 167 49* Piedmont terraces and relict valley floors, Canon de Culebras to Laguna Salada......... 169 50. Valley fill in Valle Abanico........... 180 ix Figure 51. Terrace gradients near the mouth of Camp Canyon ......................................... 52. Map of Valle Abanico fan ..................... 53* Map of Camp Canyon fan ....................... 5^• Piedmont profile near Arroyo Fagon . . . . 55, Topographic profile from the mouth of Valle Abanico to Laguna Salada ..................... 56, Topographic profile from the mouth of Camp Canyon to Laguna Salada ..................... 57* Segment to segment change in slope and overall change in slope versus distance along the range-front f a u l t ......... .. 58, Geometry of fan segmentation ................ 59* Diagrammatic sketches showing development of the Valle Abanico fan ..................... 60, Fan segmentation and complementary terrace development .................................. Plate 1, Geologic map of southern portion of Sierra del Mayor....................... Xn Page 186 192 19^ 198 200 202 208 216 221 225 pocket x TABLES Table Page X# Total precipitation at El Mayor • ••••• 37 XI. Age estimates based on soil carbonate morphology ••••••••••• 77 III. Uranium series ages of a calcareous paleosol • • • • • • • • • • • • • • • • • 109 IV. Terrace gradients near Red Ridge • • • • • 151 * V, Profile parameters for Valle Fagon, Valle Abanico and Camp Canyon fans • ......... 20b VI. ’ ’Original" segment length and segment sediment volumes ••••••• • 2±b VII. Temporal and functional correlation of pediment terraces, fan segments and lower piedmont terraces ......... •••••••• 230 ABSTRACT Sierra del Mayor, 60 1cm south of* Mexicali, Baja California, Mexico, was studied to determine the sequence of erosional and depositional events during late Quaternary time, and their relation to climatic, tectonic and eustatic controls* The following sequence of events was determined! (1) the present bedrock morphology of valleys and tribu taries of Sierra del Mayor, including the depth of valley incision, was developed prior to late Pleistocene time* (2) During late Pleistocene time the Colorado River flowed alternately north to the Salton Sink, south to the Gulf of California and west and then north around the south end of Sierra del Mayor into Laguna Salada, maintaining a large deltaic system. Uranium series ages of a calcareous paleosol at the top of deposits formed then indicate a mini mum age of 60,000 years for these materials. (3) These deposits interfinger with and are overlain by locally derived alluvium that extends up all valleys within the range. Valley filling largely post-dates the delta system. (4) Steady-state slope systems were established that extend ed from within the valleys to the basin floors. Tectonism xii disturbed the slope systems episodically, resulting in fan segmentation in the southern piedmont and pediment terrace development in the west. (5) During an ensuing interval of landscape stability, bar and channel morphology was obliterated and replaced by desert pavement on all pied mont terraces and mountain block valley floors. Dune building on lower piedmont slopes occurred during desert pavement formation. (6) Incision was initiated and a series of benches was cut into older fill. Incision has progressed to the base of the older fill in most valley mouths. Piedmont degradation accompanied valley fill removal. Throughout late Quaternary time, surface displace ment occurred episodically along faults in the southern and western piedmont. No evidence of late Quaternary fault ing was found within the mountain block. Total surface displacement along piedmont faults increases progressively eastward, and the age of the most recent activity on individual faults decreases away from the mountain front. Gentle eastward tilting of the mountain block and pied mont has occurred probably in Holocene time, as all fan- forming washes of the southern piedmont, except two that are entrenched, are on the eastern edges of their fans. Slight eastward tilting is also indicated by the sediments of the deltaic sequence. The depositional top of the sequence is approximately 10 m above sea level at the xiii eastern margin of the piedmont, and at the southwestern margin, deltaic sediments are exposed about 20 m above sea level. Deposits of the deltaic system were formed at an interval when sea level was close to present sea level. Prior to 60,000 years ago the latest stand of the sea close to that of the present occurred about 120,000 years ago, or during Sangamonian time. Xt is concluded that the deposits are remnants of a Sangamonian Colorado River delta system. Alluviation within the mountain block is either late Sangamonian or early Wisconsinan in age and an early Wisconsinan age is preferred. Alluviation was caused by climate change, which overrode the effects of base level fall along range margin faults. The nature of the climate change may have been a gradual decrease in the frequency of tropical storms. Steady-state slopes maintained nearly constant gradients throughout the remainder of early Wisconsinan time in spite of tectonic disturbance. As these gradients are a reflection of conditions within the drainage basin that are controlled by climate, it is concluded that climate did not change significantly during this interval. Landscape stability and desert pavement and dune formation occurred under conditions of extreme aridity that perhaps marked the Wisconsinan glacial maximum. Sum- xiy mer rainfall may have been nearly nil at this time, and mean annual precipitation may have been no more than 1.25 cm* Such extreme aridity prevented erosion of valley fill in valleys on the east side of the range, even though the Colorado River was then entrenched and graded to lower sea level* In the mountain block, incision following land scape stability resulted in a series of benches cut into older valley fill. At Camp Canyon, benches converge at the valley mouth, indicating that removal of the valley fill was caused by increased discharge. Discharge in creased as tropical storms gradually returned during warming at the end of the Wisconsinan Age. xv INTRODUCTION General Sierra del Mayor is 60 km south of Mexicali, Baja California, Mexico (Fig, l). The investigator's interest in this range began during a study of orbital photography of the northern Gulf of California for the National Marine Fisheries Institute (Stone, e_t al, , 1973)* 0n Gemini and Apollo photographs it was observed that a study of the alluvial geology of the range and its piedmont might allow sequences of geologic events along the margins of the Colorado River delta and Laguna Salada to be cor related, as the range is at the juncture of these two major physiographic features (Fig, 2), Subsequently, a preliminary reconnaissance was carried out involving approximately 1 week in the field. As a result of the reconnaissance it was concluded that a detailed study of the alluvial geomorphology of the southern portion of the range could solve several geologic problems of regional importance. 1 Figure 1 Location of Sierra del Mayor and boundary of the study area. 2 IK jla n o \ t n » o o ez m AlCAU V A Cotenia:, e5- K j, %T Aggftgg tunstas) ,V_,AKM A Mnr« Marrtimo Scole in Kilometers 1P\ ( \ ' MM*n«s ALIFO Boundary of Study Area ARIZONA ..SONORA CAUFORNIA LOCATION OF SIERRA DEL MAYOR 3 Figure 2 Physiographic features in the vicinity of the study area (from Thompson, 1968). Ur ^ V .— 4 •"* «~iJ J4)Dv^ ; - ar' , v^ ^ ,.«V G o i.f o C alifornia X Of? A DC D ESERT AND SURnCUNOINOS 5 Description of the Problems Along the east flank of Sierra del Mayor, a single prominent terrace that appears from a distance to be the surface of dissected remnants of alluvial fans is abruptly truncated (Fig. 3)• This feature is so striking that it was described by Kniffen (1932) as resembling a series of mine dumps; narrow, elongated, flat-topped at its outer margin, steep-sided, and abruptly terminated. Field study revealed that the material beneath the terrace is not entirely colluvial-alluvial material as reported by Kniffen (1932), but rather consists of a variety of materials in cluding alluvium, but also well-bedded sand, silt and clay of deltaic or marine origin. In addition, it was observed that the terrace continues along the eastern portion of the southern margin of the range. Westward along the southern margin of the range, the continuity of the terrace is progressively disrupted by erosion until only terrace fragments remain (Fig. k). At the range front some of the terrace fragments are pedi ments, and it is clear that multiple terraces are present. Along the western margin of the range multiple terraces are present (Fig. 5)» and the terrace surfaces are deformed by Quaternary faulting and underlain in places by older sedimentary deposits. These terraces consist entirely of surfaces of planation rather than surfaces of deposition. 6 Figure 3 Tlie single terrace along the east side of Sierra del Mayor. In the foreground is the modern flood plain of the Colorado River. 8 Figure 4, Terrace fragments of the southern pied mont. The light-toned areas, such as that in the foreground, are terrace surfaces. Figure 5* Multiple terraces of the western pied mont, Each terrace is a surface of planation. Laguna Salada and Sierra de los Cucapas are in the background. 9 On the east side of the range, the single piedmont terrace continues up valleys as a paired terrace or old valley floor and is underlain by as much as 10 m of al luvium (Fig, 6)• Alluvial fill remnants of up to 12 m in thickness are characteristic of all basins within the range and the highest fill remnant surfaces are commonly paired. In places, fill remnants sit atop low divides. Incision and redistribution of alluvium are ubiquitous in the study area. Across the mouth of Laguna Salada in Sierra de las Pintas and along the southern margin of Laguna Salada the association of morphology and materials was found to be similar to those at the Sierra del Mayor (Fig. 7)* Al- luviation, incision and terrace development are thus regional rather than local phenomena, and are therefore not the result of local faulting. These phenomena may have been induced by regional tectonism (epeirogenic uplift or basinal subsidence), eustatic sea level variations, climatic change or all of these. The problems addressed in this study are: what are the relationships of the surface morphologies and internal characteristics of sediments within and bordering the range to the ages and extent of old deltas of the Colorado River, and to climatic change, marine transgression and tectonism? 11 Figure 6. Relict valley floor in Valle del Leneros, This paired terrace is continuous with the single piedmont terrace along the eastern range front. 13 Figure 7* Piedmont terrace at Sierra de las Pintas. In the foreground are the mudflats of the modern Colorado River delta. lk Regional Significance Several observations serve to indicate the possible relationships of morphology and materials to Quaternary regional development. Kniffen (1932) believed that a terrace encircles Laguna Salada, and that the terrace edge is continuous with the 12-meter shoreline of Holocene Lake Le Conte in the Salton Basin. He noted that Brown (1923) traced this shoreline around the basin and found that it merged with the main terrace of the Colorado River. Kniffen interpreted these features to be an indication of regional uplift. Hubbs and Miller (19^8), however, do not agree with Kniffen that the edge of the terrace that borders Sierra del Mayor is continuous with the Lake Le Conte shoreline. They believed that if there were any correspondence be tween the two it would be only coincidental, and proposed that the terrace edge was produced by marine erosion dur ing a high stand of the sea that predated the last filling of Lake Le Conte. Thompson (1968) stated that stratigraphic evidence in the Yuma area indicates that the Colorado River terrace was formed by entrenchment of the river during Wisconsinan lower sea level. He proposed that the underlying materials represent aggradation during an earlier interval, perhaps Sangamonian. If this is true, the fine-grained materials 16 underlying the terrace at Sierra del Mayor may be remnants of a Sangamonian delta. At least once during the Quaternary, an incursion of waters of the Gulf of California into the Salton Basin (Downs and Woodard, 1961) provided an opportunity for marine sedimentation. Deposits along the margins of Sierra del Mayor may have occurred at this time. Near San Felipe, 130 km south of the study area, Walker and Thompson (1968) describe a sequence of Late Quaternary events involving (l) alluviation and pedimentation, (2) erosional trans gression and arroyo cutting, (3) depositional regression and arroyo filling during continued sea level rise, and recent erosional transgression and arroyo cutting. This sequence may correlate with events at Sierra del Mayor, The study area lies along or within the "boundary” (zone) separating the American and Pacific lithospheric plates, and its Quaternary development is of special tec tonic interest as the present tectonic regime probably was initiated only 4 million years ago (Larson, 1972), Defor mation of the sediments and terrace surfaces on the west side of the range was part of a Quaternary phase of the development of the Laguna Salada structural trough and may be related to movement along faults of the San Andreas system. 17 Scope and Methods In order to solve the stated problems, field mapping and surveying of the sediments and morphology of the study area was accomplished emphasizing: 1. Sedimentary structures and facies relations in the alluvium. 2. Topographic profiles of valley and piedmont terraces to determine continuity, segmentation and relation to sedimentary materials. 3. Nature and relative timing of deformation of sediments in the western part of the study area. Approximately three months were spent in the field, mainly during the fall, winter and spring of 1973 through 1975* Laboratory work included mineralogical analyses and uranium series dating of calcareous paleosols (with Dr. Kevin Knauss and Dr. T. L. Ku). These studies, together with the mapping allowed correlation, determination of environment of deposition, and distinction between locally derived sediments and those contributed by the Colorado River. Previous Work No detailed geologic mapping previously has been accomplished in Sierra del Mayor, and few references to the range exist in the literature. Work in adjacent areas 18 includes studies of the Colorado River delta (Sykes, 1937; McKee, 1939; Thompson, 1968; Meckel, 1973)> Sierra de las Pintas (Harvey and La Borde, 1968) and Sierra de las Cuca- pas (Barnard, 1968a, b). Reference to the range or ad jacent basins occurs in several reconnaissance or regional studies (Orcutt, 1890 and 1891; Jordan and Richardson, 1907; MacDougal, 1907; Cosby, 1929; Kniffen, 1932; Miller, 19^3; Hubbs and Miller, 19^8; Gastil, ej; al# , 1973)* Regional gravity maps of Biehler, eyt aT. (1964, chart l) and Kovach, e_fc al. (1962) include the study area. Useful references for regional stratigraphy are Dibblee (193^0 and Woodard (1963 and 197^0 • Acknowledgments Advice and assistance in all phases of this study were generously given by Drs. R. 0. Stone and D. S. Gors- line. In addition to the aid rendered by them in direct support of the research, their guidance and encouragement was instrumental in overcoming peripheral obstacles and enabling completion of the work. The manuscript benefitted from critical reviews by Drs. R. 0. Stone, D. S. Gorsline and R. C. McKenzie. Financial support was provided by a Union Oil Com pany of California Foundation Research Fellowship, Geo logical Society of America Penrose Bequest Research Grant 19 No* 1688-739 the Geological Society of America Quaternary Geology and Geomorphology Division J. Hoover Mackin Re search Grant for 197^* a Society of the Sigma Xi Grant-In- Aid of Research, and a National Science Foundation Graduate Traineeship. Field assistance, without which this investigation could not have been undertaken, was rendered by Gary Butler, Jeffrey Carter, Tom Carter, Roy Dokka, John Johnston, James Molnar, Michael Mulhern, James Scott, Jan Vargo and Donald Wassem. Uranium series analyses were performed by Kevin Knauss. My wife, Elizabeth, and my parents provided in valuable support and inspiration throughout the field work and preparation of the manuscript. This work was perform ed with the cooperation of the Universidad Autonoma de Baja California, Ensenada and Dr. Saul Alvarez Borrego, Director, to whom gratitude is expressed. 20 DESCRIPTION OF THE AREA Physiographic Setting The area of* study is in the central portion of* the Colorado Desert (Fig. 8). The Colorado Desert, as defined by Jaeger (1961) is a portion of the Sonoran Desert bounded on the west by the Colorado River and Gulf of California, on the east by the Peninsular Ranges and the north by the Transverse Ranges. A portion of the Colorado Desert ex tends up the Colorado River to just north of Needles, California, and here part of the area east of the Colorado River is included. The terminus of the narrow southern portion of the desert is opposite Isla Angel de la Guardia. Included in the Colorado Desert in the region of the study area are the Salton Basin, the Colorado River delta, Laguna Salada, and ranges and their associated piedmont areas to the south along the Gulf of California (Fig. 2). The rugged ranges of Sierra del Mayor and Sierra de los Cucapas, the northward extension of Sierra del Mayor, separate Laguna Salada from the delta of the Colorado River. Maximum elevations of Sierra de los Cucapas and Sierra del Mayor are 1110 and 963 m, respectively. Laguna 21 Figure 8 Boundaries of the Colorado Desert (from Jaeger, 1961)• I» ''"V U N IT E D STATES i ii — ■ « l - % GULF O F -’ /■ VC CALIFORNIA % ' 4 \ P A C in c OCEAN C 0 LORADG DES E R T 23 I Salada merges with the Colorado River delta between Sierra del Mayor and Sierra de las Pintas* On the north, Laguna Salada is separated from the Salton Basin by a low divide where alluvium and older fine-grained sediments are ex posed. The Colorado River deltaic cone, or fan-shaped flood plain radiates from an apex near Yuma, Arizona and slopes both north into the Salton Basin and south to the Gulf of California. The crest of the cone extends south- westward from an elevation of about 43 m near Yuma, Arizona to a minimum of about 11 m near Cerro Prieto 25 km south of Mexicali. To the north, the gradient of the cone is 0.8 m/km, whereas to the south a more gentle gradient of 0.35 m/km is maintained to the 5 m contour at a latitude of 32° N. Under natural conditions, this region was heavily vegetated and supported a large population of beaver (Sykes, 1937)* Below the 5-m contour a low gradient of 0.016 m/km extends to the higher-high tide line near the river mouth. Tidal influence is apparent in this region, which consists of barren, saline mud flats within which lies the estuary of the river. The mud flats continue south along the western margin of the Gulf, narrowing gradually to a terminus north of San Felipe. East of the delta is the Sonoran Mesa, an undis sected plain of alluvium that rises abruptly from the flood 24 plain and extends east to Cerro Pinacate and the ranges of Sonora. Elevation of the mesa surface ranges from 20 to 193 m, and bluffs fronting the mesa range in height from 20 m near Yuma, Arizona to 125 m near El Golfo, Sonora. Part of Laguna Salada and much of the Salton Basin are below sea level. In the Salton Basin, the surface of the Salton Sea was at an elevation of -70 m in 1968. The lowest portion of Laguna Salada is a few meters below sea level and is along the east side of the basin adjacent to the Laguna Salada and Canyon Rojo faults. Geologic Setting Stratigraphy and Lithology Sierra del Mayor, Sierra de los Cucapas, and the Peninsular Ranges in the vicinity of the study area consist of Cretaceous intrusive rocks, largely granodiorite and tonalite, and large blocks of metasedimentary rocks of un known age (Barnard, 1968). Miocene and younger volcanic rocks comprise most of Sierra de las Pintas and Sierra de las Tinajas (Gastil, <et al. , 1975)* Cerro Prieto, east of Sierra de los Cucapas, is an extrusive plug of basaltic andesite of Quaternary age that protrudes through the piedmont (Barnard, 1968). At the southeast corner of the Salton Sea, rhyolitic domes occur that are less than 50,000 years old (Doe, e_t al. , 1966) . 25 Historically active mud volcanoes are associated with both of these volcanics, and near Cerro Prieto a geothermal field has been developed. Colorado River-derived sands and muds, in addition to comprising the modern delta, floor both the Salton Basin and Laguna Salada, and interfinger with locally derived sediments that rise as complex piedmonts to the flanking mountains. In places on the piedmonts and along the moun tain margins, older, generally deformed Colorado River- derived sands, silts and clays are present (Merriam and Bandy, 1965)• These fine-grained materials interfinger with locally derived coarse grained elastics. A composite section illustrating the generalized stratigraphic rela tions in the Salton Basin-Laguna Salada-Colorado River delta region is shown in Figure 9* Oldest of the exposed river derived sediments is the Imperial Formation (Woodring, 1932) which was subdivided by Tarbet (l95l) into three members. The lowest member is 60 m thick and consists of massive and obscurely bedded gray feldspathic arenite. Above this lies 750 m of rhythmically bedded gray silty mudstone and very fine quartz arenite of the middle member. The top member is 390 m thick south of Split Mountain Gorge and is litho- logically more heterogeneous. It consists of alternating siltstone and sandstone units and intercalated massive biostromal limestone and calcareous arenite beds. A varied 26 Figure 9. Composite stratigraphic section of the Imperial Valley region (modified from Muffler and Doe, 1968; and Woodard, 1972 *) • 27 *co COMPOSITE SECTION iii co d> LU ri 2 O LU X O sand and gravel silt, sand, and clay Ocotillo Cgl. Brawley Fm. > Borrego Fm. LU LU o Canebrake Cgl. _j a. Palm Spring Fm. Imperial Fm. LU LU a. Fm. Bouse ?— Split Mountain Fm. J- LU Mecca Fm. LU LU Alverson Andesite Anza Fm. PRE-TERT1ARY Igneous and Metamorphic Rocks marine fauna is present indicating late Miocene age for some beds but a younger age for other sections (Merriam and Bandy, 1965). According to Woodard (l975)> "The lower most terrestrial vertebrate fossils known from a continuous superposed sequence of sediments that exceeds 8000 ft. (2^+00 m) in thickness (Downs and White, 1968) are associat ed with marine fossils which are common in this formation. Vertebrate faunas indicate Blancan provincial age for the uppermost Imperial beds southwest of Split Mountain Gorge.” The Imperial Formation crops out extensively in Sierra de los Cucapas south of the Canyon Rojo fault and north of Mexican Highway 2. It does not occur at Sierra del Mayor, but two isolated outcrops are present in the bajada south of Laguna Salada and opposite Sierra del Mayor (Gastil, 1975). The Imperial Formation grades upward into the Palm Spring Formation (Woodring, 1932) which consists of more than 3000 m of interbedded siltstone, claystone and arkosic sandstone. Pebble conglomerate, fresh-water lime stone, and a few thin tuff beds also are present. Sand units are generally massive, locally cross-bedded and contain concretions and petrified wood in many places. Vertebrate fossils are found throughout the section and record continuous deposition through a Blancan-Irvingtonian time interval (Downs and White, 1968). In the lower 2^+00 m of these predominantly terrestrial sediments, impoverished 29 marine invertebrate faunules provide evidence of inter mittent marine conditions within Imperial Valley as late as middle Pleistocene time (Downs and Woodard, 1961)• Both the Imperial and Palm Spring formations contain detrital Upper Cretaceous foraminifera probably derived from the Mancos Shale of the Colorado Plateau (Merriam and Bandy, 1965)* Deformed Palm Spring sediments occur along the west side of Sierra del Mayor, and undeformed younger sediments referred by the author to the Palm Spring Formation occur along the eastern, southern and western margins of the range. Overlying the Palm Spring Formation in the Borrego Badlands is light gray claystone with interbeds of sand stone, These beds have been named the Borrego Formation (Tarbet and Holman, 19***0 » and have a lacustrine fauna and a thickness of I83O m at the type locality, Dibblee (195*0 mapped similar sediments west of the south end of the Salton Sea and named them the Brawley Formation, The exposed section is 610 m thick. In the Salton Basin, the following coarse clastic sedimentary formations have been mapped: the Anza Forma tion (Woodard, 19^3 and 197*0 * the Split Mountain Forma tion (Tarbet and Holman, 19***0 , the Mecca Formation, Cane- brake Conglomerate, and Ocotillo Conglomerate (Dibblee, 195*0. All of the above are conglomeratic deposits con sisting of granitic and metamorphic debris, 30 Barnard (1968) used the term "older fanglomerates" to distinguish locally derived elastics in Sierra de los Cucapas which are equivalent to or older than the Imperial Formation. These rocks correspond in position to the Split Mountain and Mecca Formations and the lower part of the Canebrake Conglomerate. Barnard distinguished the Canebrake Conglomerate where it interfingers with the Palm Spring Formation. These rocks are exposed in places along the western and northern margins of Sierra de los Cucapas. During the course of this study, Canebrake Conglomerate was observed along the west side of Sierra del Mayor to interfinger with the Palm Spring Formation. Structure and Tectonic Development North of Sierra del Mayor, right-slip movement of varying amounts has been reported for faults with a northwest-southeast trend, and these faults are thought to be part of the San Andreas system. For the San Andreas Fault itself, right-slip movement of at least 250 km is believed to have occurred (Crowell, 1962). Seismic activity has been recorded along several of these faults, in particular the San Jacinto and Imperial faults. The range margin faults of Sierra del Mayor, ranges to the south, and the Peninsular Ranges have been the sites of large vertical displacement, but the magnitude of horizon tal movement and their relation to the San Andreas system 31 are unknown. On the basis of geodetic measurements, it has been determined that the Salton Basin is undergoing active right-lateral shear, and that the basin floor is subsiding relative to its walls (Elders, el; al. , 1972). In Laguna Salada, the position of the lowest portion of the playa surface adjacent to the east side of the basin, coupled with the presence of east—trending drainage channels tributary to this area across what was once a purely depositional surface suggest Holocene eastward to south eastward tilting of the basin floor. Numerous fault scarps cut Quaternary alluvium along the margin of Sierra del Mayor, and a fault trace transects Holocene alluvium on the east side of the Colorado River delta. The basins apparently are actively growing. The structure of this region is remarkable in that, on the basis of gravity data, as much as 6 km of low density sediments underlie parts of the Colorado River delta, Salton Basin and Laguna Salada (Kovach, at aJL. , 1962). A well drilled between the towns of Brawley and Holtville to a depth of 4098 m probably bottomed in the Palm Spring Formation (Muffler and Doe, 1968). The body of geological and geophysical data gathered in recent years indicates that the Gulf of Cali fornia Structural Province is a region of lithospheric plate separation and crustal growth (Larson, at al., 1968; 32 Moore and Buffington, 1968; Larson, 1972) which at its north end ties into the San Andreas system and at the south end connects with the East Pacific Rise, Magnetic anomaly studies at the Gulf mouth indicate that this separation began about k m.y. B.P. and has been proceeding at a rate of 6 cm/yr since then (Larson, 1972), Geo physical studies in the Salton Basin and northern gulf support separation as the origin of these basins (Elders, et al., 1972; Henyey and Bischoff, 1973)* Although these data provide a framework for the general understanding of events during the past h million years, controversies exist regarding the role of individual basement blocks, and the relationship between faults of the San Andreas system and the large tensional faults of northeastern Baja California (Gastil, 1968). Moore (1973) contended that with the initiation of spreading, parts of the eastern edge of the plate margin may have collapsed into the plate edge gap along old lines of weakness. He suggested as a possible example of such a detached block Sierra del Mayor and Sierra de los Cucapas. He also proposed that such blocks may have suffered clockwise rotation induced by right lateral shear within the San Andreas system. An additional major controversy regards the existence of a proto-gulf and an ancestral San Andreas, and the relation of either or both of these to the mountains and 33 basins of the northern gulf region. Merriam (1972) reported the presence of a major fault with possible large right lateral strike-slip displacement in Sonora, and Garfunkel (1973) proposed a model of the history of the San Andreas as a plate boundary that incorporates such an ancestral rift. Moore (1973) cited evidence in favor of a proto-gulf with a shallow, restricted marine environment in the northern Gulf Province and depths significantly greater than 1000 ra in the central portion of the province. Cenozoic History of the Colorado River Delta The growing basins of the Salton Trough-Northern Gulf Trough and Laguna Salada have been filling with debris supplied by the Colorado River since late Miocene or Pliocene time. Early deposits are the marine Imperial Formation, but the environment gradually became terrestrial from late Pliocene through middle Pleistocene time. Building of the deltaic cone into the Northern Gulf Trough accompanied by subsidence of the trough to the northwest resulted in isolation of the Salton Basin (Buwalda and Stanton, 1930). The present deltaic cone consists of Holocene deposits that extend to an undeter mined depth near Yuma (Olmsted, jet al. , 1973) and overlie transgressive sands that occur at depths generally less than 30 m (Thompson, 1968; Meckel, 1975) in the southern portion of the delta. Prior to deposition of this wedge of 3 b material, the river was entrenched, probably during the Wisconsinan maximum, about 18,000 years ago, when sea level was roughly 100 m lower than at present (van Andel, 1964)* Alternate degradations and aggradations analogous to the above, modified by tectonic processes, may have accompanied major sea level fluctuations throughout the Pleistocene* Since deposition of the Holocene deltaic cone, the Salton Basin has undergone alternate desiccation and fill ing by fresh water lakes (Hubbs and Miller, 1948; Stanley, 1963; Thomas, 1963)* The best preserved shoreline has an elevation approximately coincident with that of the present delta crest, and it is presumed that the delta was the barrier between the lake and the gulf* Climate The climate of the Colorado Desert is characterized by extreme heat in the summer months and low annual rain fall* Mean maximum temperature in July at Yuma, Arizona for nearly 100 years of record is 4l*2°C. According to Thompson (1968) local residents in the vicinity of the mud flats of the Colorado River report temperatures as high as 50°C in August and September* At El Mayor, Baja Cali fornia, about 10 km north of the study area, average yearly rainfall for the period 1950 to 1967 was 49*1 mm (Hastings, 1964 and I969K Although rainfall is low, two distinct maxima are 35 present in its distribution; a winter maximum and a late summer and fall maximum (Table l)• Winter rains occur from December through March when the Pacific Subtropical High has moved south far enough to allow low pressure areas from the Pacific to move across the desert (Jurwitz, 1953). During such circumstances large storms may move into the area, but they generally lose much of their moisture before reaching the desert. The normally gentle winter rains ac count for about 33 percent of the mean yearly rainfall. The second rainfall maximum begins in August and lasts through October, and comprises about 39 percent of the mean yearly rainfall, Hastings and Turner (1965) analyzed the complex precipitation regime for these months, and found that Baja California is affected by rainfall from three different sources! 1, West coast tropical cyclones, 2, Easterlies from the Gulf of Mexico, 3* Anticipatory rains of the winter regime. Of these, west coast tropical cyclones are the most important in the Lower Colorado Valley, and are responsible for the most violent storms that affect the region. Rainfall is quite variable. The station at El Mayor recorded brief intervals, usually in August or September, during which rainfall exceeded the yearly mean. Conversely, more than a year has passed at El Mayor with no recorded rainfall, 36 Table I* Total precipitation at El Mayor (from Hastings, 1964; and Hastings and Humphrey, 1969). Latitude: Longitude: Elevation: 32°04» 115°151 3 m Total Precipitation (mm) Station: El Mayor Agency: Secretaria de Recursos hidraulicos Year Jan Feb Mar Apr May June July .. Aug Sept Oct Nov Dec Ann 1949 0.0 0.0 0.0 0.0 0.5 0.0 11.0 1950 O • 0 3.5 O • O O • O 0.0 0.0 14.5 0.0 3.6 0.0 0.0 0.0 21.6 1951 5.0 0.0 0.0 0.0 0.0 0.0 4.1 15.0 0.0 0.0 3.5 21.2 48.8 1952 31.5 6.0 0.0 4.0 0.0 0.0 0.0 0.0 56.0 0.0 7.0 4.6 109.1 1953 0.0 0.0 11.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 11.9 1934 T 0.0 31.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 31.0 1955 14.5 0.0 0.0 0.0 0.0 0.0 0.0 100.0 0.0 0.0 0.0 0.0 114.5 1956 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1957 8.0 0.0 0.0 2.0 0.0 0.0 0.0 22.7 0.0 46.5 0.0 0.0 79.2 1958 0.0 10.3 0.2 5.1 2.0 2.0 2.0 0.4 0.0 0.0 2.3 0.0 24.3 1959 2.2 2.7 0.0 0.0 0.0 0.0 0.0 12.6 5.0 T 0.6 25.9 49.0 I960 3.6 0.0 0.0 0.0 0.0 0.0 0.0 T 35.1 0 • 0 0.0 0.5 39.2 1961 0.5 0.0 0.0 0.0 0.0 0.0 2.3 3.3 0.0 0.0 0.5 7.8 14.4 1962 6.5 0.0 5.0 0.0 1.0 0.0 0.0 1.0 9.0 0.0 0.0 49.0 71.5 1963 5.5 11.8 4.5 0.0 0.0 0.0 T 7.7 59.2 46.4 2.0 0.0 137.1 1964 0.0 0.0 0.0 0.0 0.0 0.0 12.0 1.0 0.0 0.0 0.0 0.0 13.0 1965 0.0 1.0 0.0 0.0 0.0 1.0 0.0 0.0 0.0 0.0 8.0 20.4 30.4 Table I. Total precipitation at El Mayor (from Hastings, 1964; and Hastings and Humphrey, 1969) (continued) Latitude s Longitude: Elevation: 32°04* 115°15 * 3 m Total Precipitation (mm) Station: Agency: El Mayor Secretaria de Recursos hidraulicos Year Jan Feb Mar Apr May June July Aug .Sept Oct Nov Dec Ann 1966 1967 11.0 3.0 16.3 0.0 2.5 0.0 0.0 0.0 0.0 1.4 0.0 14.5 19. 1 0.0 7.0 71.8 Mean 5.1 2.9 3.2 0.6 0.2 0.2 2.0 9.1 10.1 6.2 1.3 8.2 49.1 00 Evaporation rates in the Colorado Desert are high. An average evaporation loss of 180 cm per year in the vicinity of the Salton Sea is indicated by calculations by Hely and Peck (1964) based on energy budget and water budget considerations. Pan measurements by the Weather Bureau at Yuma show an evaporation rate of 290 cm per year, with an average monthly rate of 40 cm during the summer. Temperatures, in contrast to precipitation, differ little from year to year. Winters are mild with an oc casional frost, and summers are extremely hot. At El Mayor the coldest months are December and January, and mean temperatures for these months are 15*3 and l4.0°C, respectively. Summer temperatures regularly exceed 38°C, and the means for the monthly temperatures for July and August are in excess of 29°C for the period of record. In Ives1 (1949) opinion, "Because of low relative humidity, sensible temperatures are tolerable, and complaints of summer season discomfort are heard largely from the obese, alcoholic, and neurotic components of the population. 1 1 At Sierra del Mayor, however, high temperatures coupled with humid air brought into the area by the sea breeze regime prevalent during these months results in extreme physical discomfort. Wind regimes in the Sonoran Desert and Gulf of California have been described by Ives (l949)» Roden (1964), and Thompson (1968). Winds at intermediate eleva- 39 tions and at the surface in the broad open areas east of the gulf are controlled by interaction between the Pacific High and the Sonoran Low, and generally follow theoretical paths. Surface winds in most places, however, are not those indicated by theory but consist of the resultant of the gradient winds and the local air drainages, Xn con sequence, local winds can be determined only by direct observation. From October through March or April, in the Colorado River delta region, northwest winds predominate (Thompson, 1968). They are most persistent from October through February, with velocities of 10 to 15 knots from the direction of N, 30 to 50° W, down the gulf axis. Surface winds observed in the study area during field work, which was largely accomplished in the fall, winter and spring were highly variable. The strongest winds, however, were generally from the west or northwest. From late June through September a sea breeze system prevails and often brings muggy conditions to Sierra del Mayor, Flora The biota of the study area is characteristic of the rugged granitic ranges and flanking bajadas of the region. The interior of the range is nearly barren of vegetation. Vegetation is absent on bare rock slopes, but an occasional barrel cactus is seen on slopes with a thin ko debris veneer. Only in the lower reaches of stream chan nels is vegetation moderately abundant. Here, the larger plants are scattered desert ironwood, palo verde and smoke tree. Smaller plants are somewhat more numerous, and the most abundant of these are burro bush and brittle bush. On the piedmont, the larger plants such as those above and mesquite are confined to drainageways. Terraces on the upper piedmont are barren of vegetation, but on the lower piedmont scattered brittle bush and burro bush are present. In sandy areas the creosote bush is common. The surface of Laguna Salada is generally barren of vegetation except around the margin where halophytes grow. Along the Rio Hardy dense growths of hydrophytes occur, but away from the watercourse in the vicinity of the study area only halo phytes manage to survive. Culture ¥ithin the study area, the only cultural activities are woodgathering and mining. Woodgatherers (leneros) are frequently seen in the area. Most woodgatherers operate dual-wheeled flatbed trucks, with which they range over seemingly impassable terrain. Tracks of these trucks were seen far up bouldery canyons where the author had not dared take a 4-wheel drive vehicle. A gold mine is situated at about the midpoint of the linear southern mar gin of the range. Operation of the mine is sporadic. No 4l activity was noted during field work in 1973 o** 197^» but in April of 1975 miners were preparing to blast. During 1973 and 197^* a single family attempted to subsist by ranching on the piedmont in the southeastern corner of the study area. The harsh environment forced abandonment of the ranch in the summer of 197^* In Laguna Salada, abandoned dikes and canals attest to past failures at farm ing, and presently no attempts are being made. East of the area, on the Colorado River delta, fishing and hunting camps are maintained on the lower Rio Hardy. Fishing is excellent here, and duck hunting was good prior to the 1975 season. In 197**» a canal was con structed from the Hardy to the head of a series of barran cas draining to the northern portion of Laguna Salada. Boundaries and Accessibility The study area is bounded on the east by the Colorado River delta, and on the south and west by Laguna Salada (Figs. 1 and 2). The northern boundary was arbitrarily chosen to coincide with major watershed divides. The east side of the area is easily reached by ordinary passenger car via Mexico Highway 5* The southern and western parts of the area are reached by taking a dirt road that branches southwesterly from Highway 5 where the highway diverges from the range as it begins to cross the mouth of Laguna Salada. This road traverses the surface k2 of the mud flats until the southeastern tip of the range is reached. It then crosses a portion of the piedmont and passes back onto the mudflats to continue first west, then north up Laguna Salada, The road is passable in dry weather with a heavy-duty 2-wheel drive vehicle, provided care is taken in the sandier portions of the piedmont. Some of the piedmont is accessible by 4-wheel drive vehicle, but most of the piedmont and all of the range interior must be negotiated on foot. 43 GEOLOGY Introduction The principal concern of* this investigation is the geology of* the surficial deposits within the area delin eated in Figure 1. As morphologic and age relationships and sedimentary characteristics of the deposits are considered in detail in subsequent sections of the report, only a description of the distribution and general characteristics of the forms and deposits are given here. Because no geologic mapping had previously been accomplish ed in the range, a reconnaissance geologic map was prepared (PI. l), and a brief description of the metamorphic, igneous and older sedimentary materials that comprise the basement for the surficial deposits is included, along with a description of major structures of the range. Geomorphology The study area can be readily subdivided into three geomorphic regions (Fig. 10): the mountain block, piedmont slopes and basin floors. hh Figure 10. Geomorphic subdivisions of the study area. MOUNTAIN BLOCK PIEDMONT SLOPES BASIN FLOORS Mountain Block The mountain block, which consists of crystalline rocks, is a rugged region of steep bedrock slopes# The southern boundary of the mountain block is remarkably linear and is fault-controlled. On the west and east the mountain block boundaries are abrupt but less linear. The highest elevation in the study area, 667 m, is on the northern boundary at the crest of a slope that forms most of the steep east front of the range. This slope terminates at the south at an eastward extension of the mountain block, and is not included in the study area. Drainage is largely controlled by structure, and the trends of major drainageways are predominantly northeast-southwest (Fig. ll). Major tributaries trend northwest-southeast to form a crudely orthogonal pattern. Ridgecrests parallel the major drainageways. For the most part, the maximum elevations of major ridgecrests decreases progressively to the southeast. V-shaped valleys are characteristic, and their steep, craggy bare rock slopes comprise about 90 percent of the mountain block. Slope angles on bare rock slopes some times exceed 38°. These slopes are cut in their upper reaches by numerous chutes which direct runoff and facilitate the removal of debris. Scree slopes of various ages are present, as indi- Figure 11. Drainage basins and major drainage of the mountain block. k8 ORAiMAGE BASINS AND MAJOR DRAINAGE OF THE MOUNTAIN BLOCK \ .A e x p l a n a t i o n — BounciQ'y o f mountain blocK Dramqge divide in bedrock * • * • Drainage divide in alluvium Generalized trunk stream G e n e ra liz e d major trib u ta ry isolated bedrock ' .-A S»a • 2 K ili’ fflftftrfc cated by weathering, surface morphology and the degree of development of desert varnish. Some scree slopes are rela tively young and either are or were sites of deposition or are areas where a balance between deposition and erosion was maintained. Clasts on these slopes are fresh to slightly weathered, up to 1 m in diameter, and possess either a light coat of desert varnish or no varnish. Scree slopes of intermediate age exhibit similar clast sizes, but many clasts crumble easily under one or more hammer blows, and dark coatings of desert varnish are present. Still older scree slopes have been smoothed by weathering, creep and rainwash and do not possess desert varnish. All ages of scree are in one place or another cut by chutes. Remnants of old valley fill occur in all basins of the study area. In some basins, particularly in the western part of the mountain block, much of the fill has been removed. In most basins, incision has progressed to or slightly below the base of the fill along the length of the trunk streams. During removal of the fill, a series of generally unpaired fill strath and strath terraces were cut. A complex of relict surface forms is associated with the fill, including the oldest scree and footslopes des cribed above, rock cut slopes, alluvial fans and valley floors. In basins from Camp Canyon east, the uppermost level of valley fill is a paired terrace or old valley 50 floor (Fig, 6). The relict slope forms are graded to the old valley floors. Surface morphologies of these as sociated forms are distinctive; all possess light colored surfaces smoothed by creep, weathering, and perhaps sheet- flow. Bar and channel morphology has been obliterated on old valley floors and fan surfaces alike. Present micro relief on unreworked surfaces is generally less than 10 to 15 cm (Fig. 12). In basins west of Camp Canyon, only relict fan, rock-cut footslopes and scree slopes possess this distinctive smooth-surface morphology. All terraces lower than the upper level of valley fill in all basins exhibit bar and channel morphology (Fig. 13)* Clasts on bars of middle and higher level fill strath terraces are thoroughly weathered and are coated with dark brown to black desert varnish. Clasts often are completely rotten and disintegrate under a few hammer blows. No difference was apparent in weathering or varnish development on these terraces. On lower terraces, clasts are weathered to a depth of only a few millimeters, and exhibit light brown coatings of desert varnish. Modern channels are at or near bedrock over their entire lengths, except for channels tributary to the east ern range margin. The upper reaches are commonly incised from 1 to 3 ro, whereas a few of the lower reaches have alluvial floors. Most lower reaches expose bedrock and are at approximately the same level as the base of the relict 51 Figure 12, Microrelief on the upper level of valley fill in Valle Abanico. Figure 13* Relict bar and channel morphology on benches below the upper level of valley fill in Camp Canyon. 52 53 valley fill. Near the mouth of Canon de Culebras, a patch of the relict fill occurs in an irregularity of the bed rock floor of the modern channel. The bedrock lower reaches of the modern channels are believed to be the ex humed bottoms of the valleys that were present prior to deposition of the relict fill. Piedmont Slopes Piedmont slopes are most conveniently discussed by considering separately the eastern, southern and western range margins. On the east margin of the range within the study area piedmont slopes essentially are lacking. Only a thin strip of alluvium roughly 600 m wide lies between the mountain block and the Colorado River deltaic plain. This strip consists of small fans that emanate from minor drainageways and merge with the delta surface, and of patches of older alluvium capped by a single terrace (Fig. 3)• North of the study area in a wide embayment the pied mont is more extensive. The single terrace extends up the two main valleys of the eastern portion of the range as a paired terrace (Fig. 6). This old valley floor is coincident with the upper level of valley fill described in the discussion of the mountain block. Termination of the terrace is abrupt but irregular. If the terrace originally had a linear or curvilinear terminus, it has been highly modified by erosion. North of the study area conspicuous features of the terrace terminus are sets of parallel benches. A few benches are present in the study area, but are of ir regular occurrence. Examination of these benches disclosed that in each case they are lithologically controlled and are underlain by relatively resistant gravel beds in pre dominantly fine-grained materials. The absence of a suc cession of fluvial terraces is noteworthy, as fill strath terraces were formed during incision within the mountain block. After creation of the general configuration of the terrace margin, an escarpment was produced at the distal ends of the remnants (Fig. l4)• The escarpment, about 3 m high, cuts alluvium of intermediate age in addition to the terrace materials, but is not present in younger alluvium. The southern piedmont ranges in width from 1 to l4 km and is complex (Figs. 15 and l6). Surfaces with fan morphology comprise slightly less than one-half of the area of the piedmont. The fans extend about 1.5 km from the mountain front and have gradients that range from 3*^17° to 0.855°* From the distal ends of the fans to Laguna Salada is a complex region containing terraces of several levels, sandy plains and dunes. Elevations in the piedmont descend from a maximum of about 122 m at the mouth of Valle Abanico to almost sea level on the Laguna Salada surface over a 55 Figure 14 Escarpment in alluvium at the distal edge of terrace remnants* 56 57 Figure 15* The portion of the southern piedmont with fan morphology (middleground)• Light-toned areas in the middleground are fan terraces with desert pavement surfaces. Figure 16. Lower slopes of the southern piedmont. Dunes overlie terraces with desert pavement surfac e s. 58 59 distance of k 1cm. Surfaces in the region with fan morphology are of two types: (l) surfaces without bar and channel morphology, or desert pavement surfaces, and (2) surfaces with bar and channel morphology. At least two, and perhaps three or more levels of desert pavement surfaces are present basin- ward of Valle Abanico, separated by a total difference in elevation of only 1 to 2 m. Most desert pavement surfaces are developed on alluvium, but rock-cut surfaces with desert pavement occur at the mouths of Pediment Canyon and the adjacent small to the east. A bedrock remnant rises above the surface of the Valle Abanico fan not far from these rock-cut surfaces. Surfaces of several ages with bar and channel morphology are present, as exemplified by differences in weathering and desert varnish development. From the distal ends of the fans to the surface of Laguna Salada the piedmont is difficult to characterize, as dunes obscure much of the underlying morphology. Several levels of basin terraces can be detected, however. Like the fan surfaces, the upper terrace levels lack bar and channel morphology. These are not true desert pave ment surfaces, though, as they are primarily of a sandy nature. Bar and channel morphology is present on lower terraces, but only near fan margins. Sandy plains occur in the eastern portion and along the margin of the piedmont. They are products of aggrada 60 tion and degradation associated with the destruction of basin terraces and the transportation of debris across the piedmont. Along the piedmont margin, the plains consist of sandy fans built out onto the Laguna Salada surface. The sandy plain of the eastern portion is continuous with the largest of these fans. The basinward margin of the basin terraces is ir regular on the south and abrupt and curvilinear on the southeast. On the south, the highest terrace remnant is about h m above the Laguna Salada surface. Faint shore lines are visible on air photos but were not identified on the ground along the southwestern margin. These features do not cross the sandy fans built out onto Laguna Salada. Faint linear markings occur along the central portion of the margin and could be either shorelines or faults. The abrupt, curvilinear, southeast margin of the piedmont is an escarpment with a maximum height of about 20 m. On the Laguna Salada surface, a curvilinear feature is visible on air photos that extends from the escarpment, crosses Laguna Salada and joins a piedmont margin at Sierra de las Pintas analogous to that described above. Modern washes transect the lower piedmont slopes and are incised to depths of 2 to b m. The courses of the washes are deflected by and in some cases truncate the dune forms. The western piedmont is quite different from the 6l southern piedmont as the configuration of the surfaces has been primarily determined by faulting. The western pied mont is 3 to h km wide and is divisible into two morpho logically distinct zones with a 10-m high fault scarp as the boundary between the zones (Figs, 17 and 18), From the fault scarp to the Laguna Salada surface the morphology is simple. Gravel fans with gradients of from about 1.5 to 2.0° form a belt 0.75 Lm wide. All fans possess well preserved bar and channel morphology, and in some places clasts are darkly stained with desert varnish and thoroughly weathered. From the foot of the fans a sandy plain slopes to the surface of Laguna Salada with an overall gradient of about 0.5°* Two small areas of dunes are present on the sandy plain. Terraces of the type found in the southern piedmont are absent. From the fault to the base of the range the mor phology is exceedingly complex (Figs. 5 and 19)* Three general terrace levels (Terrace I, Terrace II and Terrace III) are present with residuals of older materials rising above each terrace. The most prominent of the residuals is Red Ridge, an arcuate remnant of hard sandstone that rises 15 to 20 m above the highest terrace surface (Fig. 20)• Granodiorite pebbles at the crest of the ridge sug gest that it is an exhumed feature. 62 Figure 17* Aerial view of the western piedmont (courtesy of R. H. Merriara). Figure 18. Fault scarp in alluvium in the western piedmont. The scarp separates a zone of simple morphology from a zone of complex morphology. The surface at the top of the scarp is Terrace III. 6 3 T i Figure 19* Terrace surfaces and remnants of older materials between Red Ridge and the range front. Photograph taken from Red Ridge. Figure 20. Red Ridge. Terrace I (right middle- ground) passes through a gap in Red Ridge and extends basinward. 65 l & t f1 Basin Floors The part of* the Colorado River flood plain in the study area is a flat surface with a sparse growth of halophytes. Surface materials are predominantly silty. The silts are saline and gypsiferous, and some areas are puffy due to the growth of these salts. Plant mounds of root-trapped sand up to 2 m in height occur at scattered localities and are the only significant relief features. The surface of Labuna Salada is barren, flat and featureless except for some of the central portion and part of the margin. A broad, shallow channel occurs in the center of the narrowest part of Laguna Salada, near the southwestern corner of the piedmont. This channel was not visited by the author. Air photo interpretation, however, indicates that the channel is less than 1 m deep and that the floor may be sandy. Along the northern margin of Laguna Salada is a zone of puffy, saline ground with scattered halophytes. Stratigraphy The geological materials in the study area can be grouped into three categories; basement rocks, older al luvium and younger alluvium. 67 Basement Rocks Metamorphic rocks Biotite schist, minor amphibolite, gneiss and chlorite schist, and rarely quartzite comprise about 25 percent of the exposed crystalline materials. The rocks are metamorphosed sedimentary rocks and neither the age of the metamorphism nor the original sedimentary age of the rocks is known. In the nearby Peninsular Range, meta- sedimentary rocks of Triassic, Jurassic and Cretaceous age have been reported (Popenoe, 195^)> and in the Transverse Ranges to the north Paleozoic metasediments occur (Bailey and Jahns, 195^0* The metamorphic rocks of the southern portion of Sierra del Mayor are different from those described by Barnard (1968) in Sierra de los Cucapas which adjoins Sierra del Mayor to the north. In Sierra de los Cucapas, marble and quartzo-feldspathic gneisses predominate, where as marble is absent and gneiss is rare in the mapped area. In the northern part of Sierra del Mayor, near El Mayor, however, quartzites are common along with biotite schist. The metamorphic rocks occur in blocks that range in size from a few square meters to 2 sq km in area. Con tacts with the surrounding granodiorite are intrusive, but no contact metamorphism is visible. In places, the con tacts are linear in gross aspect, but upon inspection 68 apophyses of granodiorite extend into the metamorphic blocks, and pieces of metamorphic rock occur as xenoliths in the granodiorite. In such places, intrusion may have occurred along pre-existing fractures. Owing to a combination of intense fracturing and weathering the rocks are weak and can often be broken apart by hand. Granodiorite The bulk of the range consists of medium- to fine grained, light gray to white granodiorite. Typically, the rock contains about 35 percent quartz, ^5 percent plagio- clase, 15 percent potassium feldspar, h percent biotite and 1 percent accessory minerals. In the eastern portion of the study area, the rock is stained red along joints by hematite, and in places this stain permeates large masses of rock. Garnet and muscovite make up as much as 5 per cent of a fine-grained leucocratic facies of the rock that occurs in some localities in the western part of the study area. A large pegmatitic body characterized by graphic granite occurs at the southwest corner of the range. Smaller pegmatite dikes are common throughout the range. The age of the granodiorite is unknown, but it is probably about the same age as similar granodiorite in Sierra de los Cucapas to the north. This age is estimated by Barnard (1968) to be about 100 million years on the basis of a 125 to 150 million year lead-alpha age of tona- 69 lite which the granodiorite intrudes. A potassium-argon age of 62.6 - 0,k million years was obtained on this tona- lite, and an age of about 75 million years was obtained on granodiorite in Sierra del Mayor by Krummenacher, e_t al. (l975)* The 100 million year estimate of the age is con sidered to be more reasonable, as the potassium-argon age may reflect the time of uplift and unroofing of the pluton (Krummenacher, e_t al. , 1975) • The rock is highly fractured to severely crushed. Joint spacing ranges from a few centimeters to as much as a meter, with fractures commonly 10 to 15 cm apart. Weathering along joints is generally minimal, and joint fillings are not common. Deep weathering was not observed, but surficial weathering to depths of 0.5 ni is in places intense. These sites of intense weathering are areas of low to moderate relief, such as isolated low hills, or at or near the sum mits of knobs in areas of high relief. In some of these highly weathered areas tafoni are common. Older Alluvium Exposures of older alluvium occur only on the west ern piedmont. These materials consist of the mildly to moderately deformed beds of the Red Ridge section, and un deformed to mildly deformed piedmont gravels that are younger than the Red Ridge sediments. 70 Red Ridge section On the west side of the range at least 300 m of mildly to moderately deformed older alluvium is exposed that in this report is informally referred to as the Red Ridge section. The base of the section is not exposed and an unknown amount of material was eroded from the upper part. During erosion, a gently sloping surface was formed that is preserved only in the northern part of the study area, and is slightly tilted to the east (Fig. 2l). The sediments have north to northeast dips of 10 to 50° east of the Red Ridge fault, and are generally horizontal west of the fault. The materials consist of a coarse-grained piedmont facies near the range that interfingers with and overlies a fine-grained basin facies. The basin facies consists of interbedded fine sand and silt which contain detrital, calcite filled foraminifera, and detrital carbonate. Feld< spar, when present, is dominated by potassium feldspar. Petrified wood occurs in places and limy layers and con cretions are common. These features are characteristic of materials supplied by the Colorado River (Merriam and Bandy, 1965)• The lack of marine fossils indicate that the deposits are terrestrial, and the combined presence of clay pebbles, parallel lamination, small-scale cross lamination, large-scale cross bedding, parting lineation, petrified wood, and intertonguing local alluvium suggests Figure 21. Tilted beds of the Red Ridge section and the eastward sloping surface that caps them (middleground)• 72 a flood plain environment. These materials are correlated with the Palm Spring Formation of Pleistocene age. The basin facies extends to the range margin be- tween Canon de Culebras and Canon Ranura, Here, and at Red Ridge, the sediments are silicified and deformed. Silicification at Red Ridge is extreme, and the sand units present are quartzites. Basinward of Red Ridge, barite veins and joint fillings, and barite crystals up to 1 cm in length, occur in a sandstone unit that is 2 m thick. Silicification and mineralization, including red coloration, is confined to zones along faults and appears to be of hydrothermal origin. Red silicified beds of Red Ridge can be traced toward the range, and are progressively less altered away from the ridge. Interfingering with and overlying the basin facies are alluvial deposits of local derivation. The largest clasts in the portions of this material equivalent to and directly overlying the basin facies are commonly of pebble size. Exceptions are in the vicinities of Canon de Cule bras and Canon Ranura, where grain size increases rapidly toward the range and cobbles and boulders are found. Above the basin facies, the local alluvium rapidly coarsens upward and boulders 2 m in diameter occur near the top of the truncated section. The local alluvium is correlated by the author with the Canebrake Conglomerate of the Salton Basin. The Red Ridge section is considered to be early Pleistocene in age, This estimate is based on the following factors: 1. The minimum age of old gravels overlying the Red Ridge section, which post-date the major deformation Red Ridge sediments. This minimum age is discussed in the next section, 2, Deposition of Palm Spring sediments began at about the beginning of Pleistocene time. Older piedmont gravels Older piedmont gravels are exposed only in the walls of washes and at the top of small, isolated residuals that rise above the terrace surfaces. They are not shown on the geologic map (PI, l). The heterogeneous nature of undeformed to mildly deformed piedmont gravels renders recognition of different ages of alluvium difficult. The best criterion for recog nition under these circumstances is the presence of pale- osols in various stages of morphogenetic development. Un fortunately, only two paleosols were found, both in the vicinity of Red Ridge, These two observations, however, indicate that older piedmont gravels of several ages occur near the surface in the western piedmont. Some may be as old as early Pleistocene, Rising above Terrace I of the western piedmont on the mountain side of Red Ridge are several residuals of 75 deformed Red Ridge sediments. The lowest of these is capped by 3 to h m of undeformed gravels. The lower 1 to 2 m contains the K fabric of a paleosol that grades upward into gravels containing disseminated caliche, Basinward from Red Ridge, the top of a zone of ex tensive caliche development occurs about 7 m below the surface of Terrace I exposed in a gully wall. The caliche zone is 6 m thick, and probably represents a compound pale osol as it consists of at least three massive carbonate- plugged horizons each 0,3 to 0,7 m thick separated by less indurated horizons each about 1,5 m in thickness. Caliche in the upper part of the zone consists of caliche blebs and disseminated caliche. Laminar horizons were not ob served. The minimum age of these gravels can be estimated on the basis of work performed by Gile, e_t aJL. (l966) in southern New Mexico on soil carbonate morphogenetic sequences. Table II is from their work. In southern New Mexico, the development of carbonate-plugged horizons does not occur on surfaces whose age is younger than late Pleistocene, This is a conservative estimate, and several factors suggest that similar profile development in Sierra del Mayor may take considerably longer than in southern New Mexico, These factors ares 1, Precipitation, The development of carbonate soil horizons depends upon rainfall infiltra- 7 6 Table II. Age estimates based on soil carbonate morphology (after Gile e£ al. , 1966, p. 3^8) Stage Diagnostic Carbonate Morphology Age of Youngest Geomorphic Surface on which Stage Occurs I Thin discontinuous pebble coatings 2600 to 5000 years II Continuous pebble coat ings, some interpebble fillings > 5000 years to latest Pleistocene III Many interpebble fillings to plugged horizon Late Pleistocene IV Laminar horizon overlying plugged horizon Mid- to late Pleistocene Thickened laminar and plugged horizons Mid-Pleistocene -a -a tion and evaporation. The higher the rate of infiltration and evaporation, the more rapid will be the growth of caliche. Rainfall in southern New Mexico averages 20 to 30 cm per year, whereas rainfall at Sierra del Mayor is only k to 8 cm per year. On this basis alone, the growth of pedogenic carbonate could be 4 to 5 times slower at Sierra del Mayor, 2, Parent material. The gravels at Sierra del Mayor consist of clasts of more acidic rocks than those in southern New Mexico, There is thus less carbonate available to form caliche, 3# Aeolian carbonate dust. Carbonate dust is con sidered by Gile, et_ aJL. (1966) to be a source for some of the pedogenic carbonate, A source for windborne carbonate dust does not exist in the vicinity of Sierra del Mayor, Considering these factors and the presence of at least three plugged horizons in the exposed gravels basinward of Red Ridge, the author believes that some of the near sur face piedmont gravels are at least as old as mid-Pleistocene and possibly early Pleistocene in age. Immediately below the caliche zone is a horizon 0,6 m thick that is indurated by iron oxides. This horizon is probably not of pedogenic origin, however, as in various places in the western piedmont lenticular zones are colored 78 red to black by iron and manganese oxides. In one loca tion, the gravels are not only colored but are indurated by these oxides. The black, hard manganese indurated lenses are in association with but distinct from the iron indurated horizons or lenses. The most highly colored and thoroughly indurated gravels occur in the vicinity of faults, and water rising along these faults is the most probable source for the oxide carrying fluids. At a locality north of the study area along the east side of the range, however, a black lenticular zone occurs in younger alluvium and does not appear to be associated with a fault. Little can be said regarding the age of the mineralization, except that it is older than Terrace III and that at least part of the mineralization is younger than the basal part of the caliche zone described earlier. The mineralized gravels are truncated by the unaltered gravels of Terrace III, and portions of the caliche in the lower part of the caliche zone have been recrystallized to calcite. Younger Alluvium Younger alluvium consists of three groups: (l) valley fill and correlative piedmont alluvium and Colorado River alluvium, (2) dune sand, and (3) alluvium derived by erosion of valley fill and piedmont alluvium. Valley fill and piedmont alluvium of the first and third groups are distinguished only as gravelly and sandy facies on the 79 geologic map. The valley fill and Colorado River alluvium and surfaces developed on the valley fill, piedmont al luvium and Colorado River alluvium are the subject of the body of the remainder of the report and are not described in detail here. Structure Mountain Block In the mountain block, the sheared nature of the rock makes recognition of faults difficult. The faults shown there (PI. l) are mapped primarily on the basis of photo interpretation, except for the portions of faults in Valle de la Paz de Dios and Valle de Leneros, where crushed and mylonitized zones were observed. Valle Fagon is a fault line valley, formed by excavation of weak materials in the crushed zone of what is probably a major fault. Westward continuation of the crushed zone was mapped by photo interpretation. It is believed that this fault connects with the western range margin fault. No indications of sense of displacement were observed on any of these faults and no evidence for movement since deposi tion of the valley fill could be discerned. Range Front Faults The southern range front is remarkably linear and 80 trends from N. 65 to N. 75 W* In spite of the linearity of the range front, a fault plane could be identified in only one place, striking N. 26 to N. 28 W. and dipping 4 5 SW. It is probable that this fault plane is only one of many within a zone along the range front. Slickensides on the fault plane are inclined 5 to 10° to the dip indicating right oblique slip for at least one episode of movement. Grooves in granodiorite of the fault plane are parallel to strike, suggesting a complex history of movement. Along the western range front, the abruptly rising mountain slopes resemble large-scale gullied road cuts, morphologically suggesting recency of formation. Slopes of the linear southern margin are not as steep, and those of eastern range front are embayed. Between Canon de Culebras and Canon Ranura, the margin fault is exposed. Movement along the fault has occurred since deposition of the Red Ridge section, as these rocks are deformed and silicified along the fault. The fault trends roughly / N. 60 ¥. and may extend into the range to Valle Fagon, al though the latest episode of movement was confined largely to the range margin. Vertical displacement of an unknown but presumably a large amount has occurred along the fault. Movement apparently has not occurred since development of Terrace I. Faults are not exposed along the eastern range front. On Gemini, Apollo, ERTS and Skylab photographs, 81 however, the linearity of the eastern front is apparent and suggestive of faulting. Piedmont Faults Movement on piedmont faults has occurred during late Quaternary time, and spectacular scarps, horsts and grabens are present in the alluvium (PI. l). A major fault of the piedmont is Red Ridge fault which strikes N. 10 W. and is best exposed at the basinward margin of Red Ridge. Ex posures of the fault plane indicate a varied history of movement. Scratches, grooves and slickensides record episodes of strike slip, dip slip, right oblique slip, and left oblique slip. The last movement was probably dip slip, as an offset cobble near the main fault plane records S cm of pure dip slip movement. East of Red Ridge fault, sediments of the Red Ridge section dip eastward at angles ranging from 10 to 50°, and at Red Ridge they are anti- clinally folded in places. Basinward from the fault the Red Ridge sediment, where exposed, are nearly horizontal. Some of these exposures may be horsts, as they appear to be slightly arched. Movement on Red Ridge fault occurred after deposition of the Red Ridge section and during deposition of alluvium that is probably the same age as val ley fill of the mountain block. Movement along this fault has not affected the surface since formation of the desert pavement of Terrace I. 82 By comparing the relative ages of materials dis placed by range front and piedmont faults and noting the distribution of surface displacement in terms of the relative ages of geomorphic surfaces, a crude pattern of development can be seen (Fig. 22). In general, the age of the last movement decreases away from the range. Although no displacement of modern sediments was observed, some of the surface displacements are quite young, probably Holocene, as they affect surfaces with well preserved bar and channel morphology. Fan morphology suggests that the range and southern piedmont are undergoing slight eastward tilting. Fan building washes of the southern piedmont, with the excep tion of two that are entrenched, occur on the east sides of fans (Fig. 23)* This movement may be quite recent and only slight as no evidence for eastward tilting can be found on the east side of the range. 83 Figure 22. Inferred age of last movement on piedmont faults. 8 k INFERRED AGE OF LAST MOVEMENT ON PIEDMONT FAULTS MOUNTAIN BLOCK E X P L A N A T IO N Pre-dates Valley Fill — ■ Post-dates Valley Fill, Pre-dotes Terroce I — Past-dates Terrace I Past-dates Terroce III , ■. Post- dat6s Surfaces With Bor ond Chonnel Morphalagy * * Uncertain, But Post - dates Red Ridge Section P IE D M O N T > \v LA G U N A SALADA 2 K lle m t l ir t Figure 23. Fans and fan forming washes of the southern range margin. 86 FANS AND FAN-FORMING WASHES OF THE SOUTHERN RANGE MARGIN EXPLANATION / / / / / Bedrock Droinoge Divide Fan - forming Wosh Fon Morgin Fan Ape* Limit of Fon Morphology FINE-GRAINED PIEDMONT SEDIMENTS Distribution Buff-colored, well-bedded to massive sand, silt and clay (Qcd of the geologic map, PI# l) is exposed beneath gravelly alluvium that caps the single terrace along the east side of the range (Fig. 2k). These fine-grained sedi ments can be traced around the southeast end of the range and are exposed in arroyos along the distal edge of the southern piedmont# In the southwestern piedmont, fine grained sediments are again well exposed# The sediments interfinger with and are overlain by gravelly alluvium and by sandy alluvium. Derivation The fine-grained sediments are not of local deriva tion but were deposited by the Colorado River. Locally derived alluvium and Colorado River alluvium are easily differentiated as Colorado River alluvium contains detrital Cretaceous foraminifers and a distinctive mineral suite (Merriam and Bandy, 1968)• There are no rocks with in the range that contain foraminifers, and locally derived alluvium is barren of such fossils# Detrital carbonate, 88 Figure 2k. Areas of exposed older Colorado River alluvium and the locations of samples and sections discussed in the text. 89 SCALE 2 KM MOUNTAIN BLOCK PIEDMONT SLOPES BASIN FLOORS PALEOSOL SAMPLE SITE SEDIMENT SAMPLE LOCALITY LOCATION OF MEASURED SECTION AREA OF EXPOSED COLORADO RIVER SEDIMENT chert and volcanic lithic fragments are common constituents of Colorado River alluvium. Chert and volcanic rocks are absent in Sierra del Mayor, and the rare occurrences of carbonate rock are restricted to the southwestern corner of the range. Three samples of the fine-grained deposits were examined for detrital foraminifers, detrital carbonate, and volcanic lithic fragments (Fig. 24). All samples con tained detrital foraminifers, detrital carbonate, and two of the three samples contained volcanic lithic fragments. In addition, along the eastern and southeastern range margins, granule and pebble-sized clasts of red and black chert and volcanic rock occur in sands and gravel lenses. Description On the east side of the range, Colorado River derived alluvium consists of massive to well-bedded silt and clay, often with ball and pillow structure, fining up ward sand units at least 6 m thick in places, and occa sional layers and lenses of gravel. A section exposed in the northeast portion of the study area is illustrated in Figure 25 and located in Figure 24. Here, the fine-grained sediments contain lenses of gravel up to 1 m thick. Granule- to pebble-sized clasts of volcanic rock and red, black and brown chert identify the gravels as Colorado River bedload, but cobble-sized 91 Figure 25. Characteristics of local alluvium and flood plain deposits. 92 CHARACTERISTICS OF LOCAL ALLUVIUM ANO FLOOD PLAIN DEPOSITS covered Local Aliuvium sond, fining upward Flood Pluin Deposits covered / sand -15 4_ sand clay with ball and pillow structure -10 2_ travel -5 100 200 40 20 VO U> clasts of granodiorite indicate significant local contri butions of detritus. The largest sand unit exposed fines upward, containing granules to small pebbles in the lower portion and silt and very fine sand in the upper part. Elongate limy nodules in the very fine sand suggest evaporation of standing or sluggish water near the sur face, Ripple cross lamination is abundant in the middle portion of the unit. Exposed beneath this unit is clay to silty clay that exhibits ball and pillow structure. The sand exposed in this section appears to represent channel fill of a large Colorado River channel incised into overbank clay and silty clay. At the top of the fine-grained sediments and be neath locally derived alluvium is a weakly developed pale- osol, which is up to 0,3 m thick. The paleosol consists of a reddish ( 5YR 5/6) oxidized zone and two to three thin to 15 mm) bands of caliche. This paleosol is present at the same stratigraphic horizon at the mouth of Valle de Leneros, but was not recognized elsewhere. At both localities, development of the paleosol at the top of a fining upward sand sequence containing evaporitic nodules near the top indicates that the paleosol was developed upon a surface of deposition rather than a surface of erosion. Southward and to the west, at the southeastern edge of the piedmont, gravel layers and lenses are not exposed, 94 but granules of* chert and volcanic rocks are present in places near the base of* sand units. Arroyos expose a sequence that consists of* massive silt and clay, and inter calated sand and silty sand, minor channels up to at least 60 m wide filled with sand, silt, clay and evaporites, and broad sand units at least 300 m wide. Figure 26 illu strates the relationship between these facies. At one locality, the margin of a broad sand unit is well exposed and is a channel fill (Fig. 27)# At the base of the exposure, which is 4 m high, granules of chert and volcanic rock are present. The unit fines upward, and is unconformably overlain by 0.5 m of sandy piedmont alluvium. In an arroyo to the east, the unit is again well exposed, and here the top of the unit can be observed. The fining upward sand is capped by silty sand and clayey silt which contains limy nodules. Overlying the unit at this locality is 2 m of sandy piedmont alluvium. In some exposures along washes in this area, slight deformation of the Colorado River sediments is evident. The presence of piedmont alluvium lying directly on the depositional top of a unit of the Colorado River alluvium at the locality described above suggests that the top of the section of Colorado River sediments is exposed here. If this is the case, then maximum uplift of the piedmont at this location since deposition of the Colorado River sediments may be only a few meters, as will be discussed 95 Figure 26, Diagrammatic sketch of Facies relations at the southeastern edge oF the pied mont . 96 c c tcw OF FACIES RELATIONS IN OIMAAMMAW pveomont s i l t , cloy, and evopornes p . |#dmont oiiuviam Piedmont terroce Massive Sand sheet Sand Modern wash C D -J Figure 27* Margin of a broad filled channel. 98 99 later. At the southwestern corner of the piedmont, Colorado River sediments occur as interbedded sandy silt and evaporite rich clayey silt, enclosing and cut by channel shaped units of fining upward, trough cross stratified and ripple laminated sand (Fig. 28), The maximum exposed thickness of a sand unit is 7 m and the maximum breadth exposed is 117 ra* The channel sands occur in a north trending zone in which individual sand units truncate other sand units. Measurements of the direction of dip of fall ing ripples in one unit are unimodal and indicate north ward flow at that locality (Fig. 29)• The maximum grain size in these exposures is medium- to coarse-sand. The finger grained Colorado River sediments occur both east and west of the channel zone. Casts of roots in growth position are common in these sediments, and together with unimodal ripple cross lamination and lack of a marine fauna indicate that the sediments are of floodplain rather than tidal flat or marine origin. All exposed Colorado River sediments of the south western piedmont discussed in the foregoing are essentially horizontal, but have been affected by faulting. The top of the sedimentary sequence is not exposed as the sediments are overlain by piedmont gravels deposited on roughly planar surfaces of erosion. 100 Figure 28, Facies relations at the southwestern corner of the piedmont. 101 FACIES RELATIONS IN THE SOUTHWESTERN PORTION OF THE PiEDMONT Sand and pebble gravel Rand and~~pea _.gra\ Sand and pea gravel Silt, clay and evaparifes Sand, caarse to medium at base, very fine at top Sand 5 r Silt, clay and evaporiles Sand Trough cross - stratified Limy nodules Scale in m eters Figure 29. Direction of dip of ripple cross lamination. 103 DiRECTiON OF DIP OF RIPPLE LAMINATION (Falling Ripples) N yi\ i ! Total Number of Observations “ 20 CROSS I Observation Correlation, Age and Implications The fine-grained sediments described are believed to be time equivalents. They exhibit a coherent deposi- tional pattern different from the pattern produced by Holocene Colorado River deposition, and distinct from the older, more deformed Colorado River sediments exposed on the west side of the range. The elements of the depositional pattern are: 1, Progressive decrease in the maximum grain size from east to west around the southern end of Sierra del Mayor; medium to large pebbles on the east side, granules to small pebbles on the south side, coarse sand on the west, 2, Broad sand units in apparently west trending channels along the southern range margin, 3, Smaller, north trending channels along the west side of the range in which flow was toward the north. This pattern suggests that the deposits are part of an ancient Colorado River delta-floodplain system (Fig, 30) in which an alluvial cone was built across the mouth of Laguna Salada analogous to the cone that now separates Imperial Valley from the Gulf of California. While such a system existed, flow of the Colorado River would have been alternately southward into the Gulf of California and west- 105 Figure 30* Flood plain sedimentation of £ Colorado River delta. former 106 GRAN DESIERTO I)EL. MAYOR SI ERR/5 Modern Flood Possible Limit of Ancient SIERRA I Plot : ! a in Kilometers 40 60 80 FLOOD PLAIN SEDIMENTATION OF COLORADO RIVER DELTA 1 0 7 ward into Laguna Salada, At this time the Laguna Salada basin would have contained a distributary channel system in which flow was to the north. When the river flowed into Laguna Salada, a lake analogous to Holocene Lake Le Conte would have been present at the north end of Laguna Salada, A minimum age for such a deltaic system can be established on the basis of uranium series analyses of caliche that occurs at the top of the deposits on the east side of the range (Table XII), Analyses were performed on caliche blebs collected at two exposures (Fig, 24), The age calculated for the caliche from exposure 1 is 40,000- years, and for exposure 2 is 60,000- years. As these are minimum ages, and the paleosols are stratigraphically equivalent, the paleosol has a minimum age of 60,000 years. The Colorado River sediments are older than the paleosol. The time that elapsed between deposition of the Colorado River sediments and formation of the paleosol is believed to be slight, however, as there is no indication of erosion prior to formation of the paleosol. The pale osol was formed on a surface of deposition rather than a surface of erosion. The Colorado River sediments and the deltaic system are thus Early Wisconsinan in age or older. Lack of appreciable deformation of the sediments in a tectonically active area suggests that if they are older than Wisconsinan, they are not appreciably older. Conditions would have been favorable for development 108 Table XII, Uranium series ages of a calcareous paleosol Sample 1: Carbonate (° / o ) U (ppm) Th (ppm) v23k ^238 Th23°/Th232 Th.2^0/U2"^^ Th23°/U23^ cor. Th232 (ppm) cor. O Age (x 10 yr.) Sample 2: Carbonate (%) U (ppm) Th (ppm) v23k/v238 Th23°/Th232 Th23°/U23^ _ 230 /tt234 Th /U cor. Th232 (ppm) cor. Age (x 103 yr.) Leach_______ Residue 76.3 0 .8 7 1 + .025 3 .0 8 6 + .1 0 8 1 .0 7 1 + .047 8 .1 3 8 + .376 1 .2 0 3 + .039 0 .8 8 8 + .036 1 .6 1 1 + .079 1 .0 3 8 + .049 0.533 + .025 0.998 + .0 5 8 0.290 11.586 -40 81.1 0.837 - .029 2.817 0 .7 4 4 i .058 8.989 1.242 i .047 0.714 2 .3 2 8 i .199 0 .9 0 8 0.487 - .032 1.31^ 0.444 1 2 .182 ^60 109 of such a system during intervals of high sea level. Dur ing Wisconsinan low stands of the sea, the Colorado River entered the Gulf of California at elevations as low as -100 m and was probably entrenched. Evidence in favor of an entrenched Colorado River is the existence of a valley system similar to the Pleistocene Mississippi Valley extending from the toe of the modern delta to a depth of 200 fathoms (van Andel, 1964). Entrenchment would have prevented the development of an alluvial cone at the mouth of Laguna Salada. If the system existed during a time of low sea level, however, then the deposits must subsequently have been elevated to their present position. Evidence within the range renders this possibility unlikely. The floors of all modern valleys within the range are incised at their mouths no more than 1 m below the base of the valley fill that overlies the Colorado River sediments. In places, patches of this valley fill occur in the floors of the modern channels. This relationship is ubiquitous, occurring in valleys tributary to the eastern, western and southern margins of the range. Base level for all these valleys is near present sea level. Had the range and flanking sediments been uplifted an appreciable amount (more than a few tens of meters), tectonic rejuvenation should have resulted in incision of the valley floors to well below the base of the valley fill. 110 If*, then, the ancient deltaic system was adjusted to a relatively high stand of* the sea, it is instructive to examine existing sea level curves to determine the most plausible time of* existence for the system. Sea level curves of Bloom, est al. (197^0 indicate that there may have been three relatively high stands of the sea in early Wisconsinan time, but that the most recent high stand ap proximated or exceeded present sea level and occurred about 120,000 years ago, or during Sangamonian time. As the deposits of this ancient deltaic system are more extensive than those of the modern one, the author considers that Sangamonian high sea level was the time development of such a large system was most likely. The development of the system was more probable during a full interglacial than during brief episodes of relatively high sea level that occurred during overall climatic deterioration. If this conclusion is correct, then all of the events to be discussed occurred largely during Wisconsinan and Holocene time , 111 VALLEY FILL Introduction All drainage basins within the range exhibit similar characteristics: dissected alluvial fill up to 12 m thick that extends to the floor of the modern channel or to with in a few meters of it. Fill extends up tributaries, and relict colluvial slopes are graded to the upper surface of valley fill. In order to understand the history and significance of this pervasive alluviation, it is necessary to consider the following questions: 1. How many episodes of filling are represented? 2. What was the relative importance of creep, debris flows and stream flow in the deposition of the materials? 3# What was the source of debris and the nature of debris production? This portion of the report is addressed to these questions, and to a consideration of timing and causes of alluviation. 112 Episodes of Filling Valley fill surface morphology is similar in all basins of the study area* Similar surface morphology, how ever, may belie complex and highly varied stratigraphic relations in the underlying fill* Determination of the presence or absence of such complexity is essential to an understanding of the history of aggradation and degradation represented by the fill* The nature of fill stratigraphic relations should be reflected by the general appearance of the fill and the presence or absence of major scour and fill structures* One of the most striking characteristics of the fill in all basins is lateral and vertical continuity and homogeneity in all exposures (Fig* 3l) • Abrupt variations in color, general grain size and the distribution of the largest clasts were not observed* This does not preclude the presence of complex stratigraphy, but is suggestive of a single episode of filling* A careful search was made for scour and fill structures* Only one structure was seen that could be interpreted as major scour and fill* In Canon de Culebras, one-third of the way downstream from the head of the canyon, a feature that is probably scour and fill structure occurs in the lower one-third of an exposure of fill 9 m in height* The apparent scarcity of major scour and fill structures 113 Figure 31, Lateral and vertical homogeneity of valley fill. The exposure is approxi mately 10 m high. 11^ can be explained in two ways: (l) the structures are in fact scarce, or (2) they are present but simply difficult to recognize. Two characteristics of the fill do indeed render scour and fill structures difficult to recognize: (1) the lateral and vertical homogeneity of the fill, and (2) the orientation of most fill exposures parallel to the valley axis. Some exposures do, however, allow a three- dimensional view of the fill, and two such exposures in Camp Canyon are especially instructive and representative of the relations observed in other valleys. One exposure provides a view of the fill at right angles to the channel axis, and also oblique to the channel axis. Scour and fill structure is visible in the section at right angles to the channel axis, but is clearly associated with a bench cut into the fill and has a maximum thickness of only 1.5 m compared to the 10 m overall fill thickness. Strati fication in the approximately 50 m long oblique section is disrupted only by minor scour and fill associated with a terrace cut into the fill. The other exposure is in the lee of a bedrock projection into the main valley. The fill is apparently uniform from the base, which is 1 m above the modern channel, to the top, which is coincident with the highest surface of fill in the canyon. On the basis of the above observations it is con cluded that valley fill in all valleys in the study area 116 was deposited during a single episode of alluviation. The fill does not consist of a series of nested fills, nor of complex, major scour and fill. The reason that major scour and fill is not commonly observed is because it is in fact rare • Depositional Processes Fill materials range from chaotic to crudely stratified and the variation in stratification occurs in a characteristic manner (Fig, 32), Deposits within and at the mouths of tributaries and at the heads of major valleys are generally chaotic (Fig, 33)• Bedding is not visible in these deposits and there is no regular variation in grain size or the distribution of large clasts. The axial por tion of the valley fill is crudely to moderately strati fied (Fig, 3h). Individual lenses or strata range from a few centimeters to a meter or more in thickness and can sometimes be traced laterally for a few meters. Scree and footslope colluvium was in one locality seen to be strati fied and to interfinger with the axial fill. Within the stratified materials, pebble and cobble imbrication is common. Only normal fluvial imbrication was observed, with the median diameter parallel to the channel axis and the dip of the clasts upstream. Chaotic debris in tributaries and the headward regions of the main valleys indicates transportation of the 117 Figure 32, Diagrammatic sketch of occurrence of stratification in valley fill. 118 DIAGRAMMATIC SKETCH OF OCCURRENCE OF STRATIFICATION IN VALLEY FILL main •'sSiey iids X x mom vcilcv floor / I Figure 33* Chaotic deposits at the head of Camp Canyon. Figure 3^* Crudely stratified deposits in Canon de Culebras. m' 7 121 material by a process that facilitated mixing; either creep or debris flows. The presence in places of strati fied colluvium indicates slow alluviation and suggests that creep may have been more important than debris flows. Fluvial imbrication in stratified axial fill indicates that although net aggradation was occurring, material sup plied to valley floors was reworked and transported by stream flow. Source of Debris and Nature of Debris Production The source of the debris is readily determined by examination of tributaries and valley sides of the major basins. Tributaries and valley side footslopes alike were sites of deposition at the time filling of the main val leys was occurring. Fill in tributaries and valley side colluvium interfinger with the axial fill of the main valley. General aggradation was occurring within the entire mountain block, and deposition in the main valleys was not due to erosion in the tributaries. The source of debris was thus the valley sides and the nature of debris production was weathering, ¥here exposed, the precipitous valley sides con tinue without an apparent break in slope beneath the col- luvial slopes and the valley fill in the valley bottoms, indicating that an extensive regolith was not available as a source for the debris. Weathering that produced the debris must have occurred during alluviation. Timing and Cause of Mountain Block Alluviation The late Pleistocene Colorado River alluvium des cribed earlier interfingers with and is overlain by local ly derived alluvium. The best exposures of this relation ship occur on the east side of Sierra del Mayor, and pro vide a basis for an estimation of the timing of mountain block alluviation. In areas where major tributaries do not occur, Colorado River alluvium abuts directly against the mountain block. The mouths of minor tributaries were flooded by river-derived sediments. Only at the mouths of major tributaries, such as Valle de Leneros, is there an appreci able quantity of local alluvium interbedded with flood plain sediments. At both northeast corner of the study area and at the mouth of Valle de Leneros, the calcareous paleosol at the top of the Colorado River sediments indicates at least a short hiatus prior to deposition of the overlying gravels. The gravels exhibit a coarsening upward sequence produced as the gravels lapped onto the old floodplain (Fig. 25)* At these two localities, the thickness of the gravels ranges from 5 to 10 m, which approximates the thickness of valley fill within the mountain block (Fig. 35)• On the basis of the presence of the paleosol and the 123 Figure 35* Relation between valley fill and flood plain deposits at Valle de Leneros• 12 k RELATION BETWEEN VALLEY FILL AND FLOOD PLAIN DEPOSITS VALLEY DEL LENEROS EX PLA N A TIO N Meters Feel 30_r-!00 200 40 0 600 800 1000 Meters caa Grovel veneer (Us! Local alluvium Old flood -plain deposits ttt paleoso l • zzzzr/ z z r tZZT-tZ7T2IZ7\ 0 '/ft; ’ o:? ">’0 2} .?»••• ?b -a- »:\ o.X - ■ -'.g -'-'. -q':- A •• t n'lir jr-rrrrn t - - inn- — t t t I . Colorado River Flood Plain 5 Segmentation may be only apparent as points are projections. similarity in thickness of the overlying gravels and fill within the mountain block, it is concluded that most of the alluviation within the mountain block occurred after deposition of the flood plain sediments. The time interval vetween cessation of flood plain deposition and the begin ning of mountain block alluviation was short, however, as there is no evidence for appreciable erosion of the flood plain sediments prior to deposition of the gravels, and the paleosol at the top of the Colorado River sediments is only slightly better developed than weak soils within the local ly derived alluvium. Two factors commonly invoked as causes for alluvia tion are base level rise, and altered conditions within the drainage basins. Base level rise can be excluded as a causative factor because: 1. Alluviation occurred in all basins of the mountain block, and the cause of alluviation should thus be applicable to all basins. Basins tributary to the western range margin were undergoing tectonic base level fall dur ing alluviation. Thus, base level rise cannot be called upon as a causative factor in these basins. 2. Deposition was occurring in trunk valleys and tributary valleys alike, from the headward regions of watersheds to valley mouths and be- 126 yond, as fan building accompanied alluviation. Observations of sedimentation induced where ponding has occurred behind dams have shown that deposition occurs only in the reach a short distance upstream from the reservoir (Leopold, Wolman, and Miller, 1964, p. 260-26l). There is no evidence to suggest that rising base level will affect deposition throughout a river system. It must be concluded that alluviation occurred due to changes within the drainage basins. Theoretically, such changes occur in several ways; a decrease in the total amount of precipitation, a decrease in storm intensity with either no change in total precipitation or an increase in precipitation, an increase in weathering rates, stripping of a preexisting regolith, and an increase in sediment yield due to overgrazing. The last two factors can be excluded at Sierra del Mayor. The other factors are all produced by climate change. 127 RELATION OF THE UPPER SURFACE OF VALLEY FILL TO SMOOTH-SURFACED PIEDMONT TERRACES Introduction Characteristics of the upper surface of valley fill are distinctive. From Camp Canyon east, bar and channel morphology is absent except in areas reworked by concentra ted flow, and desert varnish is lacking. Valley terraces lower than the upper level of fill all possess bar and channel morphology and desert varnish. Here also, with the exception of Pediment Canyon, the upper surface of valley fill is a paired terrace, or old valley floor, and relict fan slopes and scree slopes are graded to this old surface. West of Camp Canyon, valley fill is not as well preserved, and the tops of fill remnants generally have bar and chan nel morphology except for patches along their margins. The smooth-surfaced assemblage of relict fans and scree slopes is present in each valley, however, and in each case was graded to a valley floor up to a few meters higher than the present fill-tops. Surface characteristics of the highest piedmont terraces are identical to those of the upper surface of valley fill of valleys east of Camp Canyon, Lower piedmont 128 terraces, like terraces cut into valley fill, have bar and channel morphology and desert varnish. From the east side of the range, a single, continuous smooth-surfaced terrace can be traced around the southeast corner of the range and onto the southern piedmont. West along the southern pied mont, the continuity of the terrace is broken, and surfaces without bar and channel morphology occur only as patches. From Valle Abanico west, some of these patches near the range front are pediments, and from Red Ridge northward all of the smooth-surfaced piedmont terraces are cut surfaces. In addition, along the western and southwestern piedmont, at least three levels of terraces are present that lack both bar and channel morphology and desert varnish. In this section, the answer to the following question is considered! What is the relation between the smooth-surfaced assemblage of relict valley floor and slope forms in the mountain block and the smooth- surfaced terraces on the piedmont? Are these surfaces of different ages and origins, or are the surfaces closely related in time and/or origin? 129 Relation of the Upper Surface of Valley Fill and Smooth-surfaced Piedmont Terraces Developed on Alluvium Eastern and Southeastern Piedmont and Tributary Basins Along the eastern and southeastern margins of the range the relation between the highest level of valley fill and the single piedmont terrace is clear; the upper level of valley fill is continuous with the piedmont terrace* One can walk along the upper level of valley fill onto the piedmont terrace without crossing a break in slope. In places along the southeastern piedmont the terrace surface has been reworked, but pieces of the old surface remain. The old valley floor is thus equivalent to the piedmont terrace. Southern Piedmont and Tributary Basins Valle Abanico and Valle Abanico fan Valle Abanico fan occupies the central portion of the southern piedmont (Fig. 36), and covers an area of about 3*5 sq km. The fan consists of two types of sur faces: (l) patches of relatively smooth, higher ground that lack bar and channel morphology and desert varnish, and into which has been incised a parallel drainage pat tern, and (2) relatively rough surfaces with bar and chan- 130 Figure 36 Location of topographic profiles and areas of the southern piedmont mentioned in the text. 131 MOUNTAIN BLOCK SURFACES.. FAN LOWER PIEDMONT SURFACES LAGUNA SALADA FAN BOUNDARY SCALE DESER T PAVEMENT nel morphology, desert varnish, and only minor incision. The apex of the fan consists entirely of the latter surface type. About 2 sq km of drainage area Valle Abanico is tributary to the fan. Valle Abanico is oval-shaped with a constricted mouth about 0.4 km in length. From the upstream end of the mouth constriction a paired terrace lacking bar and channel morphology and desert varnish extends up the main valley and tributaries. Below this surface a flight of terraces is incised into valley fill, and each terrace possesses bar and channel morphology and desert varnish. Between the upstream end of the mouth constriction and the fan head, most valley fill has been removed, and surfaces lack ing bar and channel morphology are not present. As desert pavement surfaces are lacking on the upper part of the fan and within the valley mouth constriction, the relation between the paired terrace within the valley and the desert pavement portions of the fan surface is not obvious. Xn an attempt to determine this relationship, three radial profiles were measured with transit and stadia rod on the fan surface (Fig, 36); one along the approximate axis of the fan, and two on the west side of the fan. From the fan apex, the profiles were extended up the mouth con striction of Valle Abanico, and four points were surveyed on the most downstream remnant of the old valley floor. From the profiles (Fig. 37) » it is apparent that the old 133 Figure 37« Correlation of the highest valley terrace and highest fan terrace, Valle Abanico and Valle Abanico fan. 134 135 18 19 CORRELATION OF HIGHEST VALLEY TERRACE AND HIGHEST FAN TERRACE VALLE ABANICO AND VALLE ABANICO FAN Conyon Mouth 37 h°nne, 38 bot 36 r? 39 '°/o Meters Feet 30 n sf tr 46 * * » ( f w : > > 200 4 0 0 600 80 0 1000 _ . _l— , —1 _L~— i Feet 2 0 0 300 M«ters Scale A " ’ fan surface is segmented, with each segment consisting of a rectilinear slope. The possible origins of segmentation are discussed in a later section. For the present dis cussion, only the upper segment is of interest. Profile A-Af f passes through only the upper segment of the old fan surface, A line passing through the points on the old valley floor can be projected through the points on the old fan surface. In profiles A-A' and A-A" the same relationship is observed between the old valley floor and the upper segment of the old fan surface. In profile A-A’ it is interesting to note that the younger rubbly surface approximates the slope of the upper segment of the old fan surface. This suggests that in places the rubble is either a veneer deposited on the old fan surface, or is simply a reworked portion of the old surface, an interpretation that will be discussed more fully. Bull (l96k9 p. IOO), in his studies of segmented alluvial fans of western Fresno County, found that the slopes of valley floors and their associated active fan segments are virtually the same. His data show that the upper fan segments (active segments in his study area) and the valleys for a distance of 0.8 to 1.6 km upstream from the apex have the same general slope. That is, the gradient of a stream and of the active portion of its fan tend to attain a steady state or equilibrium slope. The relationship exhibited in the profiles strongly 136 suggests that the rectilinear upper segment of* the old fan surface was continuous with the old valley floor within Valle Abanico and that this once continuous surface was an equilibrium slope developed on the old alluvium. Thus, the old fan surfaces are not simple abandoned portions of a fan resulting from shifting loci of deposition during steady-state or quasi-steady-state fan evolution, but rather are relict surfaces of a past regime. In addition, any faulting that occurred between the range front and the old fan surface subsequent to this past regime was of in sufficient magnitude to destroy the original geometric relationships (slope and continuity) between the old fan surface and the old valley floor. Relation of the Upper Level of Valley Fill to Smooth-surfaced Pediments or Cut Surfaces Rock Fan and Barren Valley At the mouths of several minor valleys along the southern margin of the range, small pediments are well developed and their relationship to valley fill surfaces is clear. Two examples, Barren Valley and Rock Fan, are described. Barren Valley is immediately west of Valle Abanico (Fig. 36). At its mouth are two rock-cut surfaces: an upper, irregular surface of generally exposed bedrock and 137 a lower, smoother, gravel veneered surface. The lower pediment is continuous with the uppermost surface of valley fill within Barren Valley, There is no apparent change in slope between the rock-cut surface and the valley fill (Fig, 38), Rock Fan, immediately west of Camp Canyon (Fig, 36), is a fan-shaped, rock-cut surface at the mouth of a small valley. The rock-cut nature of the surface is well dis played in the incised wash that drains the valley (Fig, 39), Within the valley, the incision exposes alluvium, and the top of the alluvium, which is a paired terrace, is con tinuous with a thin veneer of alluvium on the surface of the rock fan. The continuity of the old valley floors, developed on fill, and the pediments, cut surfaces on bedrock, can be explained in any of the following manners: 1, The pediments are older than the valley fill, and the alluvium of the fill accumulated to precisely the level of the old pediment at each valley mouth, 2, The pediments are older than the valley fill and are exhumed, requiring removal of alluvium in each valley to the level of the old pediments, 3, The pediments and the old valley floors are continuous because they were formed at the same time • 138 Figure 38. Continuity of the upper level of valley fill in Barren Valley with a pediment at the valley mouth. Both photographs were taken from the same position, a* Looking up-valley from the valley mouth, b. Looking down-valley from the valley mouth. 139 1^10 Figure 39# Rock Fan; a fan-shaped rock—cut surface. i h 1 V 1^2 The first two options are dependent upon fortuitous circumstances: alluviation or degradation to exactly the level of an older surface. If the described relations were present in only one basin, such fortuity would be conceivable. As the relations are clear in two basins, and evidence presented in subsequent examples indicates that these relationships hold for all basins tributary to pied mont areas with pediment terraces, options one and two are not likely. The third hypothesis is thus favored, which requires that the old valley floors are surfaces of erosion, and implies the removal of an unknown volume of alluvium from the valleys. If this hypothesis is true, then the slightly higher, older pediment remnants at Barren Valley are either older than the valley fill, or suggest that some factor caused multiple terrace development dur ing pedimentation. Evidence favoring the latter possibil ity is presented later. Pediment Canyon A dissected, fan-shaped pediment occurs at the mouth of a valley west of Valle Abanico (Fig, 35)• The fan shaped feature consists of two terraces that lack bar and channel morphology, and a third, lower terrace with bar and channel morphology, but possessing scattered remnants of an older surface lacking bar and channel morphology. Within the valley, isolated fill remnants are present with sur 1 4 3 faces slightly rougher than those of the strikingly smooth paired terraces in the more eastward basins. Topographic profiles of the upper two pediments were made and points on the upper surface of fill were surveyed to determine the relationship between cut surfaces and surfaces on fill (Fig. 40), The two pediments were determined to be about 1 m apart in elevation at the downslope end of the upper pediment, and to converge up- slope. The upper pediment is rectilinear over the dis tance surveyed, whereas the lower pediment has an upper rectilinear segment and is gently concave upward at the lower end of the surveyed profile. The most downslope point in Figure 40 is at the crest of a remnant of alluvium that rises from the lower pediment. The two pediments are developed on both bedrock (schist, amphibolite and grandiorite) and alluvium, with the contact between the two a steeply dipping surface that is doubtless a fault. A 1- to 2-m thick veneer of alluvium rests on the bevelled bedrock. When the rectilinear slope of the upper pediment is projected up Pediment Valley, the highest remnant of valley fill lies along the projection. This remnant caps a bed rock bench. Other valley fill remnants below the projec tion are believed either to be degraded or to correlate with the third or lowest terrace. Projection of the recti linear portion of the lower pediment up Pediment Valley Ikk Figure 40. Pediments at Pediment Canyon. 1U5 146 PEDIMENTS AT PEDIMENT CANYON 3 bedrock bench approximate valley mouth 27 Meters Feet 30 T ,0° 28 26 crest of gravel residual. ?— ?______, 3 9 ~80 30 -6 0 32 -40 33 10- -20 200 400 600 800 1000 _l r - i ___L, I -.., ) Feet , ' M eters 100 200 300 Scale yields the same results as above; the projection passes through the same valley fill remnant. This suggests that the second pediment was produced by regrading the earlier pediment to a lower base level. Moreover, it is likely that the two pediments are closely related in time. Red Ridge and Canon de Culebras At Red Ridge, which extends to the mouth of Canon de Culebras, three pediment terraces are strikingly displayed (Fig. 20). The terraces are cut across deformed beds of piedmont and Colorado River alluvium of the Red Ridge sec tion, and younger alluvium that in places has been affected by faulting and associated hydrothermal alteration. The younger alluvium may be correlative with valley fill al luvium in the mountain block. The lower terrace (Terrace III) is the most extensive (Fig, 4l), and can be traced from Red Ridge northward to beyond the mapped area and from the mountain front basinward for 2.5 km, where it terminates at a fault scarp approximately 10 m in height. The highest terrace (Terrace i) is the best preserved near Red Ridge. Terrace remnants that rise above Terrace III to the north may be correlative with Terrace I. The intermediate ter race (Terrace II) is preserved only mountainward from Red Ridge, and as isolated remnants on the piedmont farther north. In the vicinity of Red Ridge, the terraces were cut 147 Figure 4l. Terraces of the western piedmont and the location of topographic profiles. 148 LOCATION OF PROFILES AND GENERALIZED DISTRIBUTION OF PEDIMENT TERRACES IN THE VICINITY OF RED RIDGE 250 5 0 0 METERS 250 To H. Canon de TI? M O U N T A IN B LO C K -Tl ? -TI? Tm >TH? \Ridge \ \ \ \ \ \ N residuals T IH T i n L O W E R P IE D M O N T To H' Laguna Salada 1 4 9 primarily by runoff from Canon de Culebras, Within Canon de Culebras, as in other valleys in the range, remnants of old valley fill 10 m or more in height occur, A transit and stadia rod traverse was made of the terraces at Red Ridge and of the fill remnants in Canon de Culebras to as certain geometric relationships between the pediment ter races and the fill tops within the valley. Terrace I Terrace I possesses a 1,2 km long rectilinear slope with a gradient of 2,319° (Table IV) that extends from the mouth of Canon de Culebras to beyond Red Ridge (Figs, 42 and ^3)• The terrace extends through a gap in Red Ridge, and prior to erosion doubtless extended through other gaps in the ridge which are at about the same elevation. Two to 4 m of alluvium cap the terrace, and the surface is a smooth desert pavement that lacks desert varnish. The terrace is not affected by the Red Ridge Fault. Two hundred meters basinward from Red Ridge, however, the rectilinear slope terminates at a fault with about 1.5 m of down-to-basin surface displacement. From here to the terrace terminus, the surface of Terrace I steps down along faults with surface displacements that range from 1 to 15 m. 150 Table IV, Terrace gradients near Red Ridge Terrace Gradient Terrace I (Profile D-D') 2.319° Terrace II 1.972° Terrace III segment 1 2.304° segment 2 1.959° 151 Figure 42. Terrace I, Red Ridge area. Profile C-C* . 152 T E R R A C E I , RED RIDGE AREA C . Residual Bedrock Slope Tributary _ to Terrace I 24? 243 244 Gravel Capped Residual Gravel Capped Residual 27 .Crest of Red Ridge Meters Feet 30- r IOOO -8 0 20- ■ Bedrock • Alluvium -60 .4 0 10- -20 200 400 600 800 1000 Feet Meters 200 100 300 Scale Figure Terrace I, Red Ridge area* Profile D-D1 . 15k RED RIDGE AREA TERRACE gravel capped residual 27 28 Crest of Red Ridge 20 22 23 24 25 26 29 30 32 34 M eters Feet 3 0 -r '00 36 37 -8 0 20- -6 0 42 43 -4 0 ’441 10- -20 46 TTEP 200 400 600 800 1000 Feet Meters 300 100 200 Scale Terrace II Terrace II (Fig. 44) cannot be identified basinward of Red Ridge, and much of the terrace mountainward of Red Ridge has been removed by erosion. The two profiles basin ward of Red Ridge are interpreted to represent Terrace I and a shallow wash cut into Terrace I. The slope of Ter race II is 1,972° and is rectilinear. Terrace III Terrace III, the most extensive of the terraces, is segmented, consisting of two rectilinear segments along the profile surveyed (Fig. 45)• The upper segment extends basinward from the mouth of Canon de Culebras for a dis tance of 700 m, and has a slope of 2,304°. From here to the abrupt, fault-controlled terrace terminus, a distance of over 1.6 km, the slope is remarkably straight, and the gradient is 1,959°* This slope is underlain by a 1 to 2 m thick veneer of unstratified alluvium that was debris in transit during the cutting of the surface. Near Red Ridge, this veneer overlies deformed beds of the Red Ridge sec tion, and in places overlies undeformed stratified alluvium with open-work structure and well developed imbrication (Fi g. 46). Basinward of Red Ridge, the debris veneer over lies alluvium that is faulted and hydrothermally altered. The greens and yellows of the hydrothermally altered allu vium contrasts with the buff to brown color of the debris veneer and emphasizes the unconformity. Relief on the un-. Figure 44. Terrace II, Red Ridge area. 157 Meters Feet -8 0 20- -6 0 -40 IQ- 200 400 600 800 1000 200 100 300 Feet Meters Scale NDGE AREA of Rec Ridge Figure 45. Terrace III, Red Ridge area. 159 TERRACE H I RED RIDGE AREA .236 245 246 Segm enl Boundary Meiers Feel . 30-plOO 20 22 23 24 25 20- -6 0 27 ■40 10- -20 2 0 0 4 0 0 600 8 0 0 1000 _ t _!__^ ll I a- Fe®' ^ 0 z 6 o So M e,e'# Scale Figure 46. Lag-gravel of Terrace III overlying stratified gravels near Red Ridge. The rock hammer in the lower part of the photograph is about 30 cm long. 161 162 conformity ranges up to 1 m, and slight channeling is evident in places. Two kinds of surface are present on this terrace. The upper segment, except for its lateral margins, and portions of the lower segment possesses bar and channel morphology and desert varnish. The distal end of the lower segment, and patches elsewhere, are desert pavement surfaces that lack desert varnish (Figs, 18 and 47). Dis tribution of bar and channel morphology is instructive be cause it extends from the upper segment onto the lower seg ment, but not across the lower segment. Rather, it extends down washes that head on the terrace. Terrace XIX can be traced around the front of Terrace I, and extends as a paired terrace up valleys incised into Terrace I, Canon de Culebras In Canon de Culebras, old fill is best preserved 0,7 km from the valley mouth and from 2,0 km from the valley mouth to the headward regions of the valley. Between these areas the valley is straight and narrow, and most of the older alluvium has been removed. In the latter region, the uppermost level of axial fill is an old valley floor that can be traced upstream the length of the profile. Ad jacent fan surfaces were graded to an older valley floor a few meters higher than the one now well preserved. Two kinds of surfaces occur on these remnants. The 163 Figure 47. The Terrace III surface (middleground). oldest fan surfaces lack both bar and channel morphology and desert varnish as do areas along the margin of the well preserved old valley floor (Fig. 48). Bar and channel morphology and desert varnish occur on most of the old valley floor. The relationship of these two surface types on the old valley floor indicates that desert pavement once covered the old valley floor but was obliterated as the present relict bar and channel morphology formed. This relict bar and channel morphology must have been present prior to the development of desert pavement when the old valley floor was an active wash surface. The group of terrace remnants 0.7 km from the valley mouth represents an old valley floor and colluvial slopes graded to that valley floor. All the surfaces are desert pavements and lack desert varnish. Figure 49 is a composite of the pediment terrace profiles and the terrace remnants within Canon de Culebras. If the upper segment of Terrace XII is projected into Canon de Culebras, it passes through the group of terrace rem nants that are 0.7 km from the valley mouth and passes through the remnant of the old valley floor that is 2.0 km from the valley mouth. The projection of the rectilinear slope of Terrace I passes above these old valley floor rem nants, and slightly below the lips of fans graded to an older, higher valley floor. Although the long profile of the valley floor rem- 166 Figure 48. Terraces in Canon de Culebras. Figure ^9* Piedmont terraces and relict valley floors, Canon de Culebras to Laguna Salada* 169 nants that extends from 2.0 to 3,6 1cm above the mouth of the valley is slightly concave upward, it is proposed that these projections provide reasonable correlation between pediment Terrace XXX and the old valley floor, and pediment Terrace I and the valley floor to which the fan remnants were graded. These correlations are based on the relation ships between the slopes of active piedmont segments and the slopes of valley floors tributary to them suggested by Bull (1964), and are consistent with relationships between old valley floors and piedmont terraces established earlier in this paper. Summary and Discussion In each valley in the eastern, southeastern and southern parts of the range, the paired terrace at the top of valley alluvium is correlative with the single smooth- surfaced piedmont terrace developed on alluvium. In the southwestern and western parts of the study area, the upper surface of valley fill is in some valleys (i.e., Barren Valley and Rock Fan) clearly correlative with a smooth surfaced pediment terrace along the range front. Where relationships are not obvious, the results of transit and stadia rod surveys indicate that this correla tion exists there also (i.e., Pediment Canyon and Red Ridge). Thus, the relationship of smooth-surfaced relict valley floors developed on alluvium and intra-valley slope 171 forms (footslopes, scree slopes and intra-valley fan shaped surfaces) to smooth-surfaced piedmont terraces is identical throughout the study area, except that multiple terraces occur in the west and southwest. From this rela tionship it can be concluded that all of the smooth- surfaced piedmont terraces were developed subsequent to valley filling. Furthermore, it was shown that the upper fan sequent at Valley Abanico was a steady-state slope in equilibrium with the slope of the surface of valley fill within the valley mouth. Examination of the slopes of the pediment terraces at Red Ridge in this light is instructive. Terrace I which is correlated here with a valley floor to which the oldest fan forms and footslopes within Canon de Culebras was graded, has a slope of 2,319°* The upper segment of Terrace III, which is correlated with the existing relict valley floor, possesses a slope of 2,304°, The difference between these slopes is only 0,015° (0° O' 54") and the slopes are here considered equivalent. The lower segment of Terrace III has a gradient of 1,959°, whereas the slope of Terrace II is 1,972°, The difference between these slopes is 0,013° (0° 01 4611), and they are also considered to be equivalent. Implied in the correlations of Terrace I and the upper sequent of Terrace III with old valley floors is that they were steady-state slopes in equilibrium with their 172 respective valley floors. The fact that the slopes are the same for the two surfaces although they are separated by a vertical distance of 10 m implies that although some factor caused terrace development, the variables controlling slope angle did not change* It is proposed that the equivalence of the gradients of Terrace II and the lower segment of Terrace III indi cates that they, too, were steady-state slopes* With this interpretation, the break in slope between the two segments of Terrace III must be hydraulic in origin. Had tilting occurred between Red Ridge and the valley mouth between the formation or Terraces I and III, or had slope angles been controlled by base level fall, the gradient equivalences cited above would not be present. Terrace II is thus interpreted as being analogous to the lower segment of Terrace III, and it is suggested that an upper segment of Terrace II was once present and has been removed by erosion. The portion of Terrace II that remains was protected from erosion by the bedrock shown in Figure kb. An analogous lower segment of Terrace I would be basinward of Red Ridge in the region affected by fault ing. Terrace II of Figure bh could be such a segment, as implied by the relations exhibited in Figure h9. If so, the gradient of this surface, which is 1.622°, would indi cate eastward (rangeward) tilting of the portion of the 173 piedmont basinward of Rod Ridge during the faulting that affected Terrace I, but before the formation of Terrace III. 174 AGE, ORIGIN AND IMPLICATIONS OF THE DESERT PAVEMENT SURFACES Relative Age Earlier, it was shown that all of the desert pave ment surfaces were formed subsequent to the episode of alluviation, A minimum relative age for formation of the surfaces can be inferred on the basis of the relationship between the distribution of desert pavement and bar and channel morphology. This relationship is best seen in Canon de Culebras and in the Red Ridge area. In Canon de Culebras, the surface of the lowest and best preserved set of paired terraces (old valley floor) possesses bar and channel morphology over most of its area, and has a microrelief of up to 0.5 m. Some areas of these surfaces, however, are desert pavement, and have a micro relief of no more than a few centimeters. Such areas generally are present along the margins of the old valley floor. These markedly different surface morphologies occur side-by-side on the same relict valley floor. An excellent example of this relationship can be seen in Figure 48. It is clear in this photograph that the area with desert pavement is part of the relict valley floor and 175 not part of a valley side footslope, nor part of a channel floor along the terrace margin. This relationship suggests that a surface with desert pavement existed prior to the development of bar and channel morphology, and that the bar and channel morphology was produced by reworking of the smooth surface. On Terrace III in the Red Ridge area, which is cor- related with the relict valley floor in Canon de Culebras, the relationship between desert pavement and bar and chan nel morphology suggests the same sequence of events. As previously discussed, this terrace extends 2.5 km basin- ward from the range front, and consists of two rectilinear segments. Bar and channel morphology occurs on the upper segment, except along the margins, and extends onto but not across the lower segment. Rather, it extends down washes that head on the terrace, and in places it terminates on the terrace. Desert pavement occurs on the distal end of the lower segment, along the eastern margin and in patches surrounded by bar and channel morphology. Subsequent to reworking of the desert pavement, much of the valley fill was removed from Canon de Culebras during a phase of incision, leaving the surface as a paired terrace both within the valley and on the piedmont. Re working of the surface is interpreted as the initial event in this episode of incision. Further evidence that this interpretation is correct 176 and can be extended throughout the study area includes the preservation of bar and channel morphology on terraces be low the desert pavement surface. Along the eastern and southeastern piedmont, and in valleys tributary to these areas, all surfaces below the single piedmont terrace and relict valley floor possess bar and channel morphology. If the piedmont terrace and relict valley floor were merely the uppermost and oldest surfaces in a sequence with com mon genesis, and desert pavement were simply an indication of elapsed time since abandonment of the surface, then successively higher surfaces should exhibit progressively better development of desert pavement. Although the degree of weathering of surface clasts increases on successively higher surfaces, bar and channel morphology is preserved on all but the relict valley floor and single piedmont ter race, and a gradational sequence of desert pavement develop ment is not evident. Thus, surfaces that possess desert pavement formed after an episode of alluviation and prior to an interval of incision. Further, an interval of desert pavement forma tion intervened between development of the surfaces and initiation of incision. Origin Desert pavement has been studied by many workers, but there is no consensus as to mode of origin. Hypotheses 177 of origin can be grouped into three categories: (l) de flation of fine particles (Chao, 1962; Clements, 1937> Symmons and Hemming, 1968), (2) removal of fines by water at the surface (Sharon, 1962; Lowdermilk and Sundling, 1950), and (3) upward migration of coarse particles (Cooke, 1970; Mabbut, 1963; Springer, 1938), An explanation of desert pavement at Sierra del Mayor must account for two characteristics of the surface: smoothness, and develop ment of the stone armor. All of the above hypotheses are attempts to explain development of the stone armor. One aspect of surface smoothness is grain size. Earlier, it was noted that clasts in modern channels are as much as 30 cm in diameter, whereas those on desert pavement surfaces are generally less than 13 cm in diameter. The question of importance is whether or not this difference was produced during desert pavement development. Figures 30 and 33 are photographs of the valley fill in Canon de Culebras, and Figure 46 is a photograph of the debris in transit during the cutting of Terrace III. The maximum size clast in these deposits is 20 to 30 cm and they are widely scattered. Most clasts are less than 13 cm in diameter. The surface of Terrace III is illustrat ed in Figure 18. The maximum size clast is 20 to 30 cm in diameter, except for isolated boulders derived from the erosion of nearby Red Ridge, and most clasts are less than 13 cm in diameter. At other places in the study area, the 178 relationship between desert pavement clast sizes and clast sizes in the underlying debris are similar. Where there are no large clasts (greater than 20 cm) on the desert pavement, there are no large clasts in the underlying materials. For example, valley fill in Valle Abanico is illustrated in Figure 50, and Figure 12 is a photograph of the desert pavement developed atop the fill. The size of the stones on the desert pavement surface is thus a func tion of clast size of the underlying materials. The breakdown of large clasts by weathering is thus not a necessary factor in pavement production at Sierra del Mayo r• When the valley floors, pediment terraces and fan surfaces that now possess desert pavement were formed, bar and channel morphology must have been present. Another aspect of smoothness, then, involves the destruction of bar and channel morphology. For this reason, the second hypothesis listed above to account for the formation of the stone armor is favored as the primary mechanism of pavement development at Sierra del Mayor as rainwash, per haps accompanied by creep, can also account for smoothing of the bar and channel morphology. Implications It was noted that the desert pavement must have formed prior to the isolation of those surfaces as ter- 179 Figure 50, Valley fill in Valle Abanico. 180 181 races. Therefore, while desert pavement was being formed there must have been insufficient runoff to maintain bar and channel morphology on valley floors and on the pied mont. The episode of desert pavement formation, then, was a time of low runoff and landscape stability. On the lower reaches of the piedmont is further evidence of the conditions prevalent during desert pave ment production. Dunes present there have a veneer of loose sand, and active slip faces are present only locally. Beneath the loose veneer the sand is compact and often friable. The dunes rest upon surfaces that possess desert pavement, but deflect and are truncated by arroyos developed during a phase of incision that followed land scape stability. It is concluded that although the dune surfaces in places appear active, the dune forms are relict and originated during the episode of landscape stability. Dune formation, coupled with runoff so light that bar and channel morphology could not be maintained on active valley floors and piedmont slopes could have been promoted in two manners: (l) the climate could have been even drier than that of the present, or (2) rainfall could have been slightly greater than or about the same as at present, but in the form of mild rains with consequent high infiltration, such as those common today during the winter months. 182 MOUNTAIN BLOCK INCISION AND THE REMOVAL OF VALLEY FILL Introduction Following the phase of landscape stability during which desert pavement and dune formation occurred, a phase of incision was initiated. During this phase, fill-strath and strath terraces were cut within the mountain block as valley fill was removed. In the piedmont, degradation was also dominant at this time. Thus, there was another change in fluvial regimen. This regimen of degradation is the most recent in evidence in the study area, and may be active today. As the factors promoting general degradation are best determined by consideration of areas within the mountain block that exhibit critical relationships, a dis cussion of piedmont degradation is reserved for a later section. General Description In each valley in the range, a flight of benches has been incised into the ancient valley fill (Fig. 13)# The benches are not paired, and in general each bench can be traced only a short distance down-valley. Bar and channel 183 morphology is present on the surface of each bench, but is best preserved on lower benches. Clasts on the highest benches are darkly stained with desert varnish, and are thoroughly weathered, disintegrating easily from a few hammer blows. Many clasts have been fractured in place, perhaps by insolation weathering (Rice, 1976), Weathering and desert varnish staining decrease in intensity from the higher to the lower benches. On the lowest benches, clasts are fresh to only slightly weathered, and desert varnish, where present, consists of only a faint stain. Incision has progressed to approximately the base of the old fill in each valley mouth. Modern channels for the most part are bedrock floored to or nearly to the valley mouth, and incision up valley has in places progressed to as much as 1,5 to 3 m below the base of the fill. Cause In Sierra del Mayor, stream incision and consequent removal of valley fill could have been initiated by any or all of the following: (l) tectonism, (2) eustatic sea level changes, (3) lowering of base level owing to incision by the Colorado River, and (4) climate change. In addi tion, each basin within the range could have been affected differently by each of these factors except for general climate change. The occurrence of the same association of 184 forms in the same position in the sequence of events for each valley in the range, however, requires a theory of origin that can be applied to the entire range* If a valley can be found, then, that exhibits characteristics that allow elimination of one or more of these factors as a cause of incision in that valley, and permits dis tinguishing the cause from among the remaining factors, then the cause of general incision and removal of valley fill within the mountain block can be determined. Such a valley is Camp Canyon. Camp Canyon is tributary to the southern range margin (Fig. 35)• A flight of benches cut into the valley fill extends up-valley to the bowl-shaped headward region (Fig. 13)* Bedrock is exposed in the modern channel to near the valley mouth, and the depth of incision into bed rock increases up-valley to a maximum of 3 m* A transit and stadia rod traverse was made up the canyon for a distance of 500 m in order to determine the relation of the benches to one another and to the valley mouth, as well as to the upper part of the fan that eminates from the valley mouth. Data gathered on the traverse (Fig. 51) show that gradients decrease from the highest to the lowest benches, and that the benches con verge at the valley mouth. Convergence of benches at the valley mouth coupled with the upstream increase in the depth of bedrock incision 185 Figure 51* Terrace gradients near the mouth Camp Canyon, < 9 4^0 35 ^ 4 \ l^,X\ TERRACE ' GRADIENTS NEAR THE MOUTH OF CAMP CANYON ® Highly weathered, dark deter! varnlth. • Slightly weathered, lltfhf deeert varnlth. ■ Modern channel (In bedrock). Meters Feet r 25 -20 6- -15 4- -10 -5 100 200 300 400 500 60 100 Scale Canyon Mouth 24 70 Upper Fan Segment demonstrate clearly that bench cutting and the removal of valley fill resulted from regrading in response to changing conditions within the watershed. Thus, causative factors originating in the piedmont or basinal areas, such as eustatic sea level fall, tectonic base level fall or inci sion by the Colorado River can be excluded. Incision initiated by any of these factors would have produced ter races that diverge toward the valley mouth, and the depth of bedrock incision should increase downstream. Two possible explanations for incision and removal of valley fill remain: (l) an increase in gradient because of southward tilting of the range with hinge at the valley mouth, or (2) increased discharge. Southward tilting of the range with a hinge at the valley mouth can be excluded for two reasons. First, evidence previously cited demonstrates that the dominant component of tilting was eastward rather than westward. Further, as the dominant direction of flow in Camp Canyon is westward, tilting of the range to the east may have served to decrease rather than increase stream gradients. Secondly, the projection of the slope of the uppermost segment of the Camp Canyon fan passes close (within 0.6 m) to the point surveyed on the highest valley fill remnant. This suggests that, as at Valle Abanico, the highest fill remnant correlates with the upper fan segment. If this is true, the geometrical continuity of this ancient equili- 188 briura surface precludes southward tilting of the range with the hinge at the valley mouth. In Camp Canyon, therefore, incision and the removal of valley fill was caused by increased discharge. As this event occurred in the same position in a sequence of events common to all basins within the range, it is concluded that a regional increase in peak discharge was the primary cause of incision and the removal of valley fill. 189 MORPHOLOGIC DEVELOPMENT OF THE PIEDMONT Introduction Climatically controlled episodes of alluviation, landscape stability and incision have been identified. After the episode of alluviation, such seemingly disparate forms as alluvial fans and pediment terraces were produced. Complicating all of the above was intermittent tectonism. The purpose of this section is to analyze the morphologic development of the piedmont in terms of these factors. Southern Piedmont Introduction The southern piedmont is the widest and is the most complex in terms of morphologies preserved. It can best be discussed using the subdivisions established earlier of a region adjacent to the range with fan morphology (upper piedmont slopes), and a region adjacent to Laguna Salada that lacks fan morphology (lower piedmont slopes). Upper Piedmont Slopes In order to understand the depositional, erosional 190 and tectonic history recorded by the fan surfaces, two approaches were used; first, maps of the surface character istics were made of two of the fans, and second, radial topographic profiles were made across the surfaces of three fans. Distribution of surface types Figures 52 and 53 are maps of the distribution of surface types on Valle Abanico fan and on Camp Canyon fan. The maps were made on the basis of air photo interpolation of field observations. Two types of surfaces were identi fied: smooth desert pavement surfaces lacking bar and chan nel morphology and desert varnish, and younger, rougher surfaces with bar and channel morphology and desert varnish. Two classes of desert pavement surface could be distinguished: unaltered to slightly modified desert pave ment, and surfaces that have been moderately to highly modified by overland flow. The maximum size clast on the unaltered surfaces ranges from 5 to 30 cm in diameter, which is in accord with the maximum size of valley fill debris exposed within the range. The maximum difference in elevation between the desert pavement surfaces is from 1 to 2 m and decreases toward the fan apex. Surfaces with bar and channel morphology were dif ferentiated into four map units at Valle Abanico and three map units at Camp Canyon, primarily on the subjective basis 191 Figure 52. Map of Valle Abanico fan. 192 MAP OF VALLE ABANICO FAN M vo u> FT ,F.T .FT- BT FT ATP \\ Dune Complex 250 \ EXPLANATION x&x&i, Bedrock Lateral Fan Llmlte — •— • Drainage J l i i. Incised Drainage Fan Surfaces Surfaces Lacking Bar and Channel Morphology; Fan Terraces | F t ] Slightly modified f;F.Tr| Highly modified Surfaces With Bor and Channel Morphology Produced by main fon- forming Channel [fl a | Youngest ED m Produced by channels heading on fan Basin Surfaces I g-j- I Basin Terraces, I_____ I BTh Is highest terrace I I I Produced by degradation ..L.i J of basin terraces Figure 53* Map of Camp Canyon fan. 19 h CANYON AN E X P L A N A T I O N £&&&& gg d ro cH c La tore! Fan Limits Surfaces Lacking Bar and Channel Morphology j F T i Slightly rnoaifiad Highly modified Surfaces With Bar and Channel Morphology i a A If.±J n r n L JLJu I in L i i i Youngest 19ST S e a l* 0 £50 500 Meters of degree of desert varnish staining. Boundaries between the units are arbitrary and indicative of gradational changes. Elevation differences between the oldest, darkly stained surface and the youngest, lightly stained surface range from a maximum of 1 to 2 m at the distal ends of the fans to less than 1 m at the apices. The youngest surface receives runoff from the modern channel. Maximum clast size on the oldest surface ranges from about 20 to 30 cm and on the youngest surface from 25 to 50 cm. Clasts on the oldest surface are thoroughly weathered whereas those on the youngest surface are fresh, Valle Abanico fan is distinctly asymmetric, with the part of the fan east of the axis being larger than the part west of the axis. The youngest surface with bar and channel morphology and the modern channel are on the east side of the fan. The Camp Canyon fan does not exhibit such marked asymmetry, but both the youngest surface with bar and channel morphology and the modern channel are on the east side of the fan. Topographic profiles Topographic profiles were made with a transit and f stadia rod down the approximate apices of Valle Fagon fan, Valle Abanico fan, and Camp Canyon fan (Fig. 35)* Each profile extended across the entire piedmont and onto Laguna Salada, but only the part crossing the fan is discussed 196 here. Profile traverse lines were chosen to cross the oldest surfaces preserved on the fans. It was anticipated that these data would aid in understanding conditions in both the piedmont and the drainage basins during the earliest phase of fan history recorded by surface morph ology. / At Valle Fagon fan only one piedmont terrace is present, and the entire topographic traverse was across this surface. Subdued, relict bar and channel morphology occurs on the upper l/3 of the profile. The remainder of the terrace surface is smooth. Both Valle Abanico fan and Camp Canyon fan have several closely spaced terrace levels. At both fans the profiles were surveyed on the highest of these surfaces. As at the Valle Fagon fan, the upper parts of the profiles cross surfaces with bar and channel morph ology, whereas the lower parts cross smooth desert pavement surfaces. The topographic profiles are presented in Figures 5^, 55 and 56, which also show the distribution of desert pavement and bar and channel morphology. Each fan profile consists of three segments rather than a smooth curve. With the exception of the Valle Abanico fan, each segment is rectilinear. The middle seg ment of the Valle Abanico fan has a rectilinear upper part and a convex-upward nose. Table V lists the profile parameters for each fan. The slope for each segment was obtained from the profiles 197 / Figure 5^# Piedmont profile near Arroyo Fagon. 198 > A FAN SEGMENTS P IE D M O N T PROFILE N E A R A R R O Y FAGON Bar and Channel Desert Pavement Morphology 20 25 30 35 40 43 dune' Meters Feet 30 H- 50 F.an Limit 24 20 20- F 20 -3 0 F22 F 2 1 f 23 -20 10- \ 30 -10 32 33 Laguna 200 400 600 800 1000 Feet 1 — 1 1 -1 * Meters 100 200 300 J' Salada Scale Figure 55. Topographic profile from the mouth of Valle Abanico to Laguna Salada. 200 53 FANI SE G M EN TS Desert Pavement 54 Bar and ChanneL Morphology ,55 TOPOGRAPHIC PROFILE FROM MOUTH OF VALLE ABANICO TO LAGUNA SALADA 56 57 58 59 6 0 62 63 64 65 66 73 67 72 ,68 . 6 9 M eters Feet 74 75 76 77 78 - 8 0 79 80 89 82 83 20- 84 85 - 6 0 Fan Lim it - 4 0 IQ - 92 100 93 94 95 96 97 98 -20 103 l o s ^ ^ cluneSjOl 200 4 0 0 600 800 1000 i | ____i i ____i reeT 100 2 0 0 300 Me,ers 107 106 dum 108 109 110 dune 1 1 2 Laguna Salada 117 118 119 120 Figure 56. Topographic profile from the mouth of Camp Canyon to Laguna Salada, 202 L FAN S E G M EN TS Bar and Chanr^L Morphology Desert Pavem ent TOPOGRAPHIC 25 20 28 23 £ 29 M eters Feet 30 " T 100 30 Fan Lim it - 8 0 20- - 6 0 -40 1 0 - -20 200 400 600 800 1000 Feet 300 Meters 100 200 TOPOGRAPHIC PROFILE FROM MOUTH OF CAMP CANYON TO LAGUNA SALADA 33 dune 37 crest of gravel ridge dune 4 0 42 dunes' 43 4 4 dune 52 53 ’ ’ ’ du"ne\54 38 39 5 8 47 gravel capped ridge 48 4; 67 62 66 63 64 gravel capped ridge Laguna 70 Salada Scale f Table V. Profile parameters for Valle Fagon, Valle Abanico and Camp Canyon fans Segment Length _ (ml . . . No. of data points Gradient (deg.) Dif. from grad, of seg. 1 Dif. from grad, of seg. 2 / 1 366 12 1.345 Valle Fagon fan overall slope 2 287 9 1 .2 0 6 0.139 — 1.04° 3 793 25 0.855 0.490 0.331 1 909 10 3.417 Valle Abanico fan overall slope 2 433 10 3.013 o.4o4 - 3.23° 3 268 3 2.603 0.814 o.4io 1 215 4 3.005 Camp Canyon fan overall slope 2 7k7 11 2.383 0 .6 2 2 — 2.38° 3 213 4 1 .9 2 2 1 .0 8 3 o .4 6 l ro o - p - in the following way: A line defining each segment was drawn through the data points with a straight edge. Only for the middle segment of the Valle Abanico fan was interpolation necessary. For all other segments, no data point lies more than 50 cm off the line defining the seg ment, and for three of the nine segments the line passes through each data point. The line defining the segment was then projected to its intersection with a vertical line passing through the initial point of the traverse (fan apex), The vertical fall and horizontal distance from this intersection to the last data point on the segment was then measured and the slope calculated, Origen of segmentation Fan segmentation was recognized by Bull (1964), who speculated that each segment represented a period of stability in the history of the fan, Hooke (1965, p* 121) pointed out that, "Implicit in Bull1s discussion of seg mented fans is the assumption that the depositional slope of a fan is uniquely determined by conditions in the source area. Thus, the fan and its source area appear to con stitute a steady-state system with respect to slope as well as fan area," For any fan, this steady-state system can be dis rupted by changing the size of discharges, the mean size of the sediment, the sediment concentration in the flows 205 or by tectonism (Hooke, 1967, 1968 and 1972). Complicating interpretation is the fact that different types of distur bance can produce similar changes of fan morphology. For example, if the slope of the fan is decreased by tilting and conditions within the drainage basin remain unchanged, a new fan will be built out over the head of the old fan to create a new fan segment that would have a slope equal to the original slope of the old segment. Conversely, during a period of tectonic quiescence, if a decrease in discharge within the drainage basin causes alluviation and an in crease in slope, a new, steeper fan segment would be built out over the head of the old fan. Additionally, an in crease in slope due to tilting, or an increase in dis charge could cause a new fan segment to be built out from the toe of the old fan. A variety of changing combinations of tectonic and climatic conditions can be envisioned that would produce a complex history of segmentation. Some seg ments could be buried by later fan deposition, and thus be lost to surface study. What, then, does the visible seg mentation of the fans at Sierra del Mayor mean? Examination of Table V and the topographic profiles reveals several things about the segments: 1. Segment boundaries generally do not coincide with the boundaries between desert pavement and bar and channel morphology. 206 2. There is no fan to fan relation between segment length and position on the fan. Segments can not be correlated on the basis of relative length. 3. Changes in slope from segment to segment on one fan are not the same as the segment to segment changes on any other fan. Segments cannot be correlated on the basis of equal slope changes. 4. The amount of the segment to segment change in slope increases from east to xirest, although there is no such east-west change in overall fan slope or individual segment slope. This last relationship is intriguing, and is the key to understanding the origin of segmentation for these fans. Figure 37 is a semilogarithmic plot of the segment to segment change in slope and overall change in slope for each fan versus the position of the fan along the range front fault. Position on the range front fault is measured t relative to Valle Fagon. For example, Valle Abanico is / 3.73 km west of Valle Fagon. It is obvious that there is a direct correlation between the difference in slope from segment to segment on these fans and position along the range front fault. Hooke (1963), in a field and laboratory study of alluvial fans, found that the steady-state slope of a fan is controlled by the calibre of debris supplied to the fan 207 Figure 57# Segment to segment change in slope and overall change in slope versus distance along the range-front fault. 208 Segment to segment slope change in de gre e s SEGMENT TO SEGMENT CHANGE IN SLOPE VS. POSITION ALONG RANGE-FRONT FAULT Camp Canyon 1 .9 V a lle Abanico 8 .7 6 5 V a l l e Fagon Segment .4 .3 2 0 3 2 4 5 6 D istance in km from Valle Fagon along range - front fault 209 and by the nature and magnitude of the debris transport ing processes. These factors are determined within the drainage basin. If fan segmentation resulted from changes in the steady-state slope of the fan due to changes within the drainage basin, then it follows that if the change in slope between segments is related to position along the range front, then drainage basin characteristics, indivi dual segment slope and overall fan slope should be related to position along the range front. It is clear from Table V that there is no correlation between overall fan slope or segment slope and position along the range front. It must be concluded that segmentation resulted from tectonism. Additional evidence favoring a tectonic interpretation occurs in Camp Canyon and Valle Abanico, Near the valley mouths, bedrock is exposed in the modern channels, which are incised no more than 2 m below the highest fan surfaces. At Camp Canyon, the projected inter section of fan segment 3 with the range front occurs at a depth of about 8,7 m, at Valle Abanico it occurs at a depth of 10,7 m, If the third segments of these fans were once steady-state slopes adjusted to conditions within their drainage basins, then uplift of the drainage basins rela tive to the fan apices has occurred. Discussion By combining information contained in Table V and 210 Figure 57 with, the concept of a steady-state fan slope, it is possible to reconstruct the nature and magnitude of tectonic activity affecting this portion of the piedmont during fan segmentation, and to assess the stability and character of conditions within the drainage basins. Evidence has been presented that indicates that the upper fan segments at both Valle Abanico and Camp Canyon represent relict steady-state slopes that were in equi librium with the slope of the upper surface of valley fill within each valley. No topographic data exists for Valle / Fagon, but field examination indicates that this relation ship also is present there. If the other fan segments were also equilibrium slopes, and conditions within the drainage basins during formation of these segments were the same as those during formation of the upper segment, then the original slope of each fan segment should have been the same as the present slope of the upper fan seg ment • The present slopes of segments 2 and 3 of each fan can be explained by assuming that they were originally steady-state slopes that were affected by two episodes of movement along the range front fault, accompanied by slight northward (rangeward) tilting of the piedmont. The amount of tilt, although small, increased logarithmically eastward along the range front during each episode. The amount of increase was less during the first episode than during the 211 second. The total northward component of tilting of the piedmont was 1,083° at Camp Canyon, 0,8l4° at Valle Abanico and 0.490° at Valle Fagon, Between each phase of movement, tectonic conditions were stable enough to allow new equi librium slopes to be built out over the heads of the old fans • It is not possible to ascertain the amount of dis placement at the range front due to tilting of the pied mont versus the amount due to simple vertical movement. An approximation of the total amount of displacement at the range front, however, is given by the intersection of the projection of the segment with a vertical projection of the range front. These amounts are: at Camp Canyon, 8,7 m, t at Valle Abanico 10,7 m, and at Valle Fagon, 4,4 m. It is interesting to note that although the amount of tilting of the piedmont increases toward the east, the amount of dis placement does not. This can be interpreted to indicate either a warped axis of tilt, which is closest to the range front at Camp Canyon, or that absolute uplift of the range (independent of tilting of the piedmont) was greatest at Valle Abanico, In this regard, it should be noted that the mouth of Valle Abanico is 122,9 m above the Laguna Salada surface, compared to 83*8 m for Camp Canyon and / 44,8 m for Valle Fagon, One characteristic of the segmentation is the ap parent lack of correlation between segment lengths and any 212 other parameter. If the "original" segment length is used rather than the visible length, that is, the visible length plus the projected length to the range front, there is still no correlation with overall fan slope, segment slope or position on the range front fault (Table Vi). Especially interesting is the fact that there is no fan to fan correlation of segment lengths of analogous seg ments, or of the volume of sediment necessary to build up the surface to the steady-state slope (Table Vi). If the fan materials were deposited during a regime of deposition, then the segments, which were shown to be time correlative, should possess lengths that are time dependent and reflect the volume of sediment dejjosited. As conditions within the drainage basins were demonstrated to have remained essentially constant, then the debris supply to each fan must have remained essentially constant. Because the fan shaped sediment wedge underlying segment 1 of each fan was produced during the same period of time, the ratio of the volume of this wedge to the volume of the underlying wedge of segment 2 should be the same for each fan, as the sedi ment wedges of the 2nd segments were also deposited during equal time intervals. As can be seen in Table VI, this ratio is not constant from fan to fan. An explanation for these apparent inconsistencies can be found by considering the geometry of fan segmenta tion and the geologic conditions at Camp Canyon and Valle 213 Table VI. "Original” segment length and segment sediment volumes. Seg ment Original length (m) Sediments volume (m ) Volume ratio Valle Pagon fan 1 366 43,953 2 671 2,584,610 0.02 3 1415 Valle Abanico fan 1 909 1,699,135 2 1317 3,759,597 0.k5 3 1561 Camp Canyon fan 1 215 60,897 2 951 583,655 O H • O 3 1143 214 Abanico. Figure 58 illustrates the geometry of segmenta tion for segments 1 and 2 of any fan assuming no buried segments and a vertical mountain front. It can be seen from the figures that C, the length of segment 1, is dependent upon , the steady-state slope of the fan, ©, the angle of segment 2 (which is determined by the amount of tilt of the piedmont), and by the distance T. At Camp Canyon and Yalle Abanico, bedrock is exposed in the modern channel near the range front, and the distance T is entirely a product of displacement along the range front fault. As , the steady-state slope of the fan, is con stant, then the distance C is determined solely by T and ©, both functions of tectonism. Once the steady-state slope was established at these fans, the segment length and the volume of sediment in the fan shaped wedge remained constant; that is, segment length and sediment volume were time independent. The steady-state slopes were thus slopes of transportation and were not produced in response to general conditions of al- luviation. Thus, alluviation within the mountain block had ceased prior to formation of the 3^d segments of the fans, the oldest surfaces preserved in the region of fan morphol ogy. Two important implications can be drawn from these considerations. First, the volume of sediment in a fan segment cannot be used to estimate rates of deposition un- 215 Figure 58. Geometry of fan segmentation. 216 GEOMETRY OF FAN SEGMENTATION Range Front Fault Original condition T + (d sin 8 ) 8. After movement along fault, tilting of pied m an and building of new steady - state slope less the segment surface is known to be an aggrading sur face rather than a surface of transportation. Second, pedimentation and deposition can occur contemporaneously on the piedmonts of adjacent drainage basins, depending upon which response is necessary to produce an equilibrium slope. For example, Pediment Canyon is between Camp Canyon and Valle Abanico. The range front fault is a short distance beyond the mouth of this canyon, and pedi mentation occurred here between the fault and the valley mouth during the time that fan segmentation was being produced at Camp Canyon and Valle Abanico. From the above discussion the following conclusions can be summarized: 1. Segmentation preserved on the highest fan ter races resulted from a northward component of tilt of the piedmont during movement along the range front fault. 2. The segments present on each fan are time cor relative, i.e., the second segments of each fan were produced during the same time interval. 3. For an individual fan, the original slope of segments 2 and 3 was the same as the present slope of the upper fan segment, and represents the steady-state slope for that fan during the period of segment formation. 218 4. Segment length and the volume of sediment deposited to build up the steady-state slope were determined by the geometry of tectonism and were time independent. 5. Movement along the range front fault ceased prior to production of the upper fan segment. 6. Alluviation within the mountain block had ceased prior to initiation of the fan segments preserved on the piedmont. 7. During production of fan segments, conditions within drainage basins remained essentially constant. A slope of transportation was maintained that was adjusted to remove from the drainage basin the debris that was supplied from the valley side slopes. 8. Either deposition, pedimentation or landscape stability can occur under these conditions, depending upon the response necessary to achieve or maintain an equilibrium slope. This response may vary from basin to basin within the same region. Fan evolution Information obtained from analysis of the topographic profiles can now be combined with inferences derived from the distribution of surface types and with morphologic 219 relationships found within the range to describe the over all development of fan morphology of the southern piedmont. Figure 59 is a diagrammatic representation of the inferred sequence of events at Valle Abanico fan. All of the events of this sequence occurred after the phase of alluviation within the mountain block described earlier. The sequence can be exjjlained as follows: A. Development of segment 3 "by building of a steady-state slope at an angle of 3.^17° that encompassed the entire fan. B. Fan morphology after formation of segment 1. The fan has now been tilted northward by 0.8l4° in two increments, an initial tilt of 0.4l0° and a second tilt of 0.4o4°. Segment 1 was built by deposition from the break in slope between segments 1 and 2 to the valley mouth. Down fan from this break in slope, segment 1 may have extended as a surface of erosion cut into seg ments 2 and 3> thus isolating them as fan ter races. C. Sometime after the formation of segment 1, removal of valley fill within the range began. Earlier, it was shown that desert pavement formation occurred prior to the development of bar and channel morphology, and prior to the initiation of incision within the range. The 220 Figure 59. Diagrammatic sketches showing development of Valle Abanico fan. p 21 DIAGRAMMATIC SKETCHES SHOWING DEVLOPMENT OF VALLE ABANICO FAN - 7 / FT FT m t FT FT iS l FT 222 oldest surface with bar and channel morphology on the fan is interpreted as heralding the beginning of incision and removal of valley fill within the range. The first stage of this episode on the fan consisted of deposition of a convex lobe of debris at the fan head (Fig. 37) and transportation of debris across most of seg ment 1. The desert pavement surface of segment 1 thus was largely obliterated, but where deposition was not too great, the surface was merely reworked and the slope was maintained. In Figure 5^+» the upper portion of segment 1 has been reworked to produce bar and channel mor phology, the lower portion has not. D, E, and F. Continued removal of valley fill, combined with fan degradation near the apex and deposition along the east side and near the foot of the fan. ”F" represents the present fan surface. Arroyos heading on the fan and sub sidiary fans at the mouths of the arroyos were developed during this interval. Lower Piedmont Slopes Most of the area of the lower piedmont is obscured by dunes and by sandy plains formed during the destruction / of older surfaces. Only at the foot of the Valle Fagon fan 223 are the lower piedmont slopes well exposed. Here, the lower piedmont consists of three closely spaced terraces that converge toward the mountain and are incised by modern washes (Fig. 5k)• The maximum vertical separation between the highest and lowest terrace is about 2 m at the escarp ment that forms the margin of the piedmont in this area. Separation between the terraces decreases gradually toward the range until the surfaces lose their identify about 790 m from the piedmont margin. The terrace surfaces are smooth desert pavement surfaces but are noticeably sandier than the fan surfaces. Maximum clast size ranges from 6 to 10 cm. The piedmont terraces can be interpreted in two ways: as a result of base-level fall or as complements of the three fan segments. In either case, formation of the terraces was induced by tectonism. If the idea proposed for the origin of the fan seg ments is correct, terraces similar to these would be required in order to transport debris to the basin floor across the lower piedmont. Figure 60 illustrates the sequence of formation of such terraces in conjunction with fan segmentation. Initially, two slopes would be present, a steady-state fan slope adjusted to conditions within the drainage basin, and a lower piedmont slope adjusted to transport the fine debris carried beyond the fan toe. In response to northward tilting of the piedmont, two new 22k Figure 60. Fan segmentation and complementary terrace development. 225 FAN SEGMENTATION AND COMPLEMENTARY TERRACE DEVELOPMENT Fan \ S \ \ 3 Lower Piedmont A. Initial condition 3 B. After one episode of piedmont tiltina v \ \ u > v \ C. After two episodes of piedmont tilting i l l slopes would be formed; a new steady-state fan slope and a new piedmont slope. Presumably, each new surface would possess the same angle of slope as its predecessor. The piedmont slope would extend into the fan area to its junc tion with the new fan segment. With a second episode of tilting, a third fan segment and a third piedmont slope would be formed. Recognition of these slopes in the region of fan morphology would be extremely difficult as, at the / slope angles of the fan and lower piedmont at Valle Fagon, maximum relief between surfaces would be on the order of 1 m. Unfortunately, topographic data are too few to adequately evaluate the hypothesis that the terraces were produced in conjunction with fan segmentation. If the terraces were correlative with the fan segments, they should possess the same angular relationship as the fan segments. That is, the angular difference between Ter race I and Terrace III should be 0.490°. The available data indicates that the angular difference is only 0.127°. Only two data points, however, are available for calcula tion of the gradient of the lower terrace, and three for the intermediate terrace, so the slopes of these surfaces and the angular difference between the terraces cannot be considered accurate. The concept that the piedmont terraces correlate with fan segments could possibly be evaluated with addi- 227 tional surveying in combination with grain size analyses. This hypothesis requires that as for each fan segment, each piedmont terrace should be a steady-state slope and should have had similar original slope angles and grain size parameters. The alternate possibility, that the terraces were produced by base level fall, cannot be ruled out. With such an origin, the terraces could be either the same age as or younger than fan segments, but older than surfaces with bar and channel morphology. Lower piedmont slopes below Camp Canyon appear to / be similar to those below Valle Fagon, but are obscured by dunes (Fig. 56). Terraces are present, but the surfaces are not well enough exposed for slope determination. They could be explained by either hypothesis proposed above. Beyond the toe of the Valle Abanico fan, the dune- obscured terrain (Fig. 55) is quite different from that previously discussed. Arcuate scarps are present and slopes are irregular. There is an abrupt basinward drop of 10 m across the dune nearest the range. Such surface ir regularities are most likely explained by subsidence. Relative age of the dune forms on the piedmont was considered in the discussion of the origin and implication of desert pavements. Dunes rest upon all terrace surfaces and are therefore younger than terrace formation. Modern 228 washes transect the dunes, and the dunes are older than the modern washes. Western Piedmont The essential elements of the morphology of the western piedmont have been discussed, and it was proposed that the gradients of the piedmont terraces could be ex plained in terms of equilibrium slopes. It is now pro posed that Terrace I and the upper segment of Terrace III are analogous to fan slopes of the western piedmont, that is, in equilibrium with the gradients of the lower reaches of their tributary valleys. Also, it is proposed that Terrace II and the lower segment of Terrace III are analogous to the lower piedmont terraces of the southern piedmont, With this interpretation, the pediment terraces of the western piedmont were induced by two episodes of move ment along the Red Ridge fault during a time when condi tions within the drainage basins remained essentially constant. The Red Ridge fault is continuous with the range front fault of the southern piedmont along which movement occurred to produce fan segmentation (Fig, 22). The correlations listed in Table VII between pediment terraces, fan segments and lower piedmont terraces are therefore suggested. After formation of Terrace III, the western piedmont 229 Table VII. Temporal and functional correlation of pediment terraces, fan segments and lower piedmont terraces Pediment Terraces Fan Segments Lower Piedmont Terraces Terrace III Temporal Correlation 1 lower Terrace II 2 middle Terrace I 3 upper Terrace III upper segment Functional Correlation 1 lower segment lower Terrace II middle Terrace I 3 230 was truncated by faulting. This event occurred prior to removal of valley fill from the range, as arroyos carved into the fault scarp provided corridors for the transporta' tion of debris from the valleys. Fans were built out from the mouths of the arroyos during removal of the valley fill, and the fans and sandy plains below them are graded to the floor of Laguna Salada. Eastern Piedmont Within the study area, the eastern piedmont is not wide enough to allow evaluation of the sequence of events determined for the southern and western piedmont areas. North of the study area, however, air photo interpretation reveals similar fan and lower piedmont morphologies to those present along the southern piedmont, and the same sequence of events is presumed to have occurred. 231 ORIGIN OF THE TERMINUS OF SMOOTH-SURFACED TERRACES Introduetion Smooth-surfaced terraces of the piedmont are not graded to the modern basin floors. In the east, the ter race terminus is irregular, but abrupt, and generally near the piedmont-floor plain boundary. The southeastern ter race terminus is a curvilinear escarpment at the mud flat margin. Along the piedmont-playa border in the southern part of the study area the end of the terraces is irregular and obscured by sandy fans and plains. On the west, Ter race III ends abruptly at a linear scarp about 0.75 km from the playa. These different characteristics suggest dif ferent origins, and therefore each area is considered separately. Southeastern Terminus The southeastern terrace terminus is a curvilinear escarpment whose top ranges from 18 to 20 m above the sur face of Laguna Salada. Of this height, 10 m is the escarp ment itself and the remainder is a gentle slope which falls from the base of the escarpment to Laguna Salada over a distance of about 220 m. The latter slope consists of fans 232 at the mouths of modern washes and sand aprons between. Fan slopes are not in equilibrium with the slopes of washes (Fig. 55)> indicating that the washes were truncated and that the base of the escarpment was buried. Youthfulness of the escarpment is indicated by the lack of equilibrium between the slopes of fans and washes and by two other features. First, the escarpment can be traced northward where it diminishes in height and occurs as a truncation of the distal edge of the embayed and dis sected eastern terrace terminus. Second, on air photos, this curvilinear feature can be traced across Laguna Salada to the northern tip of Sierra de las Pintas, where it forms the northern margin of the inactive portion of the Holocene tidal flats. Although visible on air photos, this feature cannot be detected on the ground in the central part of Laguna Salada. At La Corvina, 10 km southeast of the study area, the modern mud flats overlie a transgressive sand that occurs at a depth of about 10 m below the surface (Thomp son, 1968). Meckel (1975) reported that this sand is generally present in the mud flats, and rests upon oxidized, burrowed, Pleistocene mud flat deposits. It is proposed that the curvilinear escarpment at Sierra del Mayor, and the curvilinear feature visible on air photos that crosses Laguna Salada, represents a Holocene shoreline that marks the limit of Flandrian trans 233 gression. The shoreline is buried by recent muds where it crosses Laguna Salada, and beach deposits at the base of the escarpment at Sierra del Mayor are buried by mud flat ac cretion and by modern fans and sand aprons. Continuation of the escarpment northward indicates that transgression progressed an unknown distance up the Colorado River valley. Studies by Thompson (1968) of the modern mud flats indicate that they formed by depositional regression near modern sea level. Radiocarbon ages of shore lines in the southern part of the tide flats show that the modern mud flats were well developed by 3000 years B,P, (Thompson, 1968), The escarpment at Sierra del Mayor is therefore older than 3000 years. The sea level curve of Curray (1965) depicts a rapid Holocene sea level rise until about 3000 years ago, when sea level attained approximately its present position. As the top of the transgressive sand at La Corvina occurs at a depth of about 10 m, which was the approximate position of sea level about 7000 years ago (Curray (1965)* it is probable that the sea cliff at Sierra del Mayor was cut when sea level was not far below its present position, and is on the order of 3000 to J000 years old. Along its northern limit, the escarpment truncates all but the youngest of terraces produced during incision and removal of alluvium from the mountain block. By middle 234 Holocene time, then, most of the alluvium had been removed from within the mountain block. Southern Terminus A definite escarpment separating the smooth sur faced terraces of the southern piedmont from the surface of Laguna Salada is not present. The piedmont-playa junction in most places is a zone of sandy fans and plains derived from reworking of piedmont alluvium that masks the edge of the smooth-surfaced terraces. Where the terrace edge is visible, it is a broad, slightly dissected, gently sloping zone of erosion that is approximately 300 m in breadth. Sediments exposed in this zone predominantly are sandy, gypsiferous, locally derived alluvium. The difference in elevation between the Laguna Salada surface and the distal edges of the nearest flat topped terrace remnants ranges from 3 to h m. On the south side of Laguna Salada a similar piedmont margin is present. The age and origin of this erosional margin and the physical appearance of this portion of Laguna Salada can be surmised by considering the implications of the buried shoreline that crosses Laguna Salada from Sierra del Mayor to Sierra de las Pintas. If, during Wisconsinan low stands of the sea, erosion had removed sediments of Laguna Salada to much below present sea level, then during Holocene trans gression, Laguna Salada would have formed a long arm of the 235 sea, and the piedmont margin should be recognizable as a shoreline. Throughout Wisconsinan time, this portion of* Laguna Salada must have been a sandy alluvial plain whose eastern boundary was the valley of* the entrenched Colorado River, After abandonment of the Laguna Salada portion of the Sangamonian Colorado River floor plain, the processes most responsible for shaping the basin floor must have been deposition of prograding piedmont sands on the flood plain alluvium, and tectonism. The role of tectonism can be estimated by examining the topographic profile of Figure 56 and the pattern of faulting for this area illustrated in Figure 22. The irregular nature of the terrain and arcuate pattern of faulting from the northernmost dune toward the basin is interpreted as the result of sub sidence, Similar arcuate patterns outlining irregular features on the modern Laguna Salada floor are visible on air photos. It is probable that subsidence was episodic throughout Wisconsinan time. There are then two possibilities for the origin of the terrace terminus. First, it is a Wisconsinan erosional feature developed because of differential movement between the basin and the piedmont-mountain block complex, perhaps during rangeward tilting of the piedmont. Subsidence basin- ward of this boundary to slightly below modern sea level created the avenue that allowed spring tides and flood 236 stage waters of the Colorado River to enter Laguna Salada after Holocene depositional regression had built up the surface of the Colorado River delta. Second, it is a Holocene erosional feature induced by removal of basinal alluvium through the agencies of spring tides and flood stage waters of the Colorado River. The first hypothesis is preferred, as the major function of the flood waters was mud flat accretion rather than erosion. Additionally, arroyos that transect the boundary either were formed or were in existence during the early stages of removal of mountain block valley fill, as alluvium was transported across the piedmont in these washes. The margin was therefore in existence |?rior to mountain block incision and the removal of valley fill. Western Terminus The origin of the linear, 10-m high western terrace margin is clear; it is a fault scarp. The fault can be traced southward into a broad zone in the piedmont where multiple horsts and gravens are present. The scarp cuts the youngest of the pediment terraces, and is therefore younger than these features. Arroyos had been developed across the scarp prior to incision and removal of valley fill within the mountain block, as arroyos heading on the terrace received debris transported across the terrace at this time. Inception of the scarp occurred, then, during 237 or just before the episode of desert pavement formation. Eastern Terminus The origin of the eastern terminus, adjacent to the Colorado River flood plain, is the most difficult to deter mine as the area may have been affected by tectonism, eustatic sea level variations, and aggradations and degra dations of a major river. Accordingly, all of the follow ing mechanisms or combinations, thereof, should be consider ed as possible origins for the terminus: 1. An original depositional feature. 2. Marine erosion. 3. Faulting. k. Erosion by the Colorado River. An Original Depositional Feature A possible origin for the terminus is that it was always present. That is, the terrace ends abruptly because the terrace materials are deltaic deposits that formed rapidly in a body of water, and the terminus corresponds to the maximum position of steeply dipping foreset beds. This hypothesis is easy to evaluate, as the materials be neath the terrace surface consist of a few meters of gently inclined and locally derived alluvium overlying Colorado River flood plain alluvium (Figs. 25 and 35)* Both of these depositional units are truncated at the terrace 238 terminus. The terminus is not, therefore, a primary depositional feature. Marine Erosion Hubbs and Miller (19^+8, p, 110) proposed that the truncation of the piedmont around Sierra del Mayor was due to marine erosion that occurred prior to the last filling of Lake Le Conte, They stated (p. 111), "Indications of a recent temporary rise in Gulf level extend from San Felipe to Cerro Prieto, The fresh marginal truncation of the al luvial apron, causing it to look like a long row of mine dumps, obviously represents marine erosion at a recent higher-than-present Gulf level," The shape of the terrace margin on the east side of Sierra del Mayor, however, precludes formation by marine erosion. Marine erosion in unconsolidated sediments would have produced a straight or curvilinear shoreline (compara ble to the southeastern terrace margin) except where bed rock buttresses were present. The shoreline would not have conformed, to the shape of alluvial fans, as does the ter race terminus, but rather would have cut across the toes of the fans as trangression progressed. Moreover, spits, bars, beaches and dunes are absent along the base of the terrace bluffs, and the remains of marine organisms do not occur inland from the reach of historical high tides. 239 Faulting At its terminus, the terrace surface ranges in height from 18 to 25 m above the surface of the modern flood plain, and the top of the Pleistocene flood plain sediments is about 15 ni above the modern flood plain. Only a few meters of uplift is thus necessary to ascribe the origin of the terrace and the initiation of the terrace terminus to movement along faults. Evidence was presented earlier that demonstrated episodic uplift of the range accompanied by faulting. It is possible to account for the elevation of the ancient flood plain and overlying fan deposits entirely by uplift and to attribute the origin of the terrace terminus to faulting. Morphologic features and deposits beyond the study area, however, indicate that faulting was not the only factor involved in terrace development. Along the east side of Sierra de las Pintas, a single piedmont terrace borders the modern mud flats (Fig. 8)• The similarity be tween this terrace and the terrace at Sierra del Mayor is striking, and the height of the distal edge of the terrace above the mud flats is roughly 18 to 25 m. From Yuma, Arizona upstream 2^1-0 km to Parker, Arizona along the Colorado River, an extensive river ter race occurs at heights of from 20 to 25 m above the flood 2h0 plain (Olmsted, et al. , 1973» Metzger, ejfc aJ., , 1973)* Metzger, e_t al. (l973» P* G-32) point out that in the Parker-Blythe-Cibola area little or no structural activity has occurred since formation of the terrace deposits, and imply that the terrace and later erosion and deposition can be related to changes in the level of the Gulf of Cali fornia. According to Olmsted, et al. (l973> P* H-27), 11. . . the terraces may have been formed at the last major higher sea-level stand during the Sangamon Interglacia tion. 1 1 Pleistocene tidal flat deposits occur beneath the modern Colorado River mud flats (Meckel, 1975) and Pleis tocene tidal flat deposits that may be continuous with these are exposed in sea cliffs north of San Felipe (Thomp son, 1968), Thompson calculated, on the basis of sedi mentary structures in the exposed deposits, that sea level at the time of deposition was about 7 "to 8 m above that of the present, if the deposits have not been uplifted. This is approximately the elevation determined for maximum Sangamonian sea level (Ku, e_t al. , 1974). Thompson stated that the deposits may be part of a Sangamonian delta of the Colorado River. Arguments supportive of a Sangamonian age for the Pleistocene flood plain deposits at Sierra del Mayor have already been presented. If the interpretations of Olmsted, et al. (l973)> Thompson (1968) and the author are correct, 241 the 20 to 25 rn terrace along the Colorado River, the Pleistocene flood plain deposits at Sierra del Mayor and the Pleistocene tidal flat deposits north of San Felipe are correlative, and represent parts of a Sangamonian Colorado River flood plain-delta system. Without any uplift at all, then, a terrace would be present today along the east side of Sierra del Mayor, simply because the flood plain-delta system was graded to a higher sea level than presently obtains. If no uplift had occurred along the east side of Sierra del Mayor, the height of the top of the Pleistocene flood plain sediments above the modern flood plain should be from 7 to 10 m. The top of the sediments is, however, 15 m above the modern flood plain, indicating from 5 to 8 m of uplift since deposition. It is concluded that the height of the terrace surface above the modern flood plain results from a combination of uplift and the original al titude of deposition of the flood plain sediments. Erosion by the Colorado River The Rio Hardy impinges on Sierra del Mayor 5 km north of the study area, and it is obvious that minor trimming of the piedmont has been accomplished by the Colorado River and its distributaries during the Holocene. The shape of the terrace terminus clearly reveals, though, that Holocene erosion by the river was not responsible for 242 its origin* Configuration of the terrace margin closely conforms to the shape of alluvial fans issuing from the range* Where fans are not present, the terrace margin is deeply embayed* The terminus neither conform to the shape of meanders of the Colorado River, nor has it been straightened owing to erosion by distributaries of the Colorado River* The main activity of the river during the Holocene along the east side of Sierra del Mayor has been deposition and the building up of the modern flood plain. During the Wisconsinan, the Colorado River was en trenched and graded to various low stands of the Gulf of California. The depth of incision was possibly as much as 100 m during the Wisconsinan maximum. Entrenchment of the Colorado River is the most likely cause for the initial truncation of Sangamonian flood plain and Wisconsinan pied mont deposits. The truncated flood plain deposits would have been exposed to erosion except where protected by caps of coarse alluvium derived from Sierra del Mayor. Erosion of the soft, fine-grained flood plain sediments is far more rapid than is erosion of the coarse local detritus. Differential fluvial erosion of the flood plain sediments easily ex plains the correspondence of the shape of the terrace margin to the shape of the fans that issue from Sierra del Mayor, and explains the embayments that are present be tween sources of coarse alluvium. 2^3 The age of* inception of this margin can be esti mated relative to events already described* Washes that head on the terrace and that post-date the terrace margin were in existence prior to mountain block incision and removal of valley fill. These washes were avenues of transport for alluvium across the terrace surface during the initial stage of valley fill removal. The terrace terminus is therefore older than the phase of mountain block incision and removal of valley fill. SUMMARY OF LATE QUATERNARY HISTORY AND A CLIMATIC INTERPRETATION Summary The late Quaternary history of the southern portion of Sierra del Mayor can be summarized as follows: 1, Valleys in Sierra del Mayor were cut to their present depths prior to Wisconsinan time, 2, During the Sangamonian Interglaciation, the Colorado River maintained a deltaic system in which the river flowed alternately into the Gulf of California, the Salton Basin, and around the south end of Sierra del Mayor into Laguna Salada, Flood plain deposits laid down at this time girdle the south end of the range, 3, Valleys within the mountain block were allu- viated after deposition of the Colorado River flood plain deposits, Alluviation was accom plished by the slow accumulation of debris rather than by stripping of a pre-existing regolith. The height of valley fill remnants averages about 10 m. Deposition was initiated by climate change. Steady-state slopes in equilibrium with the slopes developed on valley fill in the lower reaches of valleys were formed on the southern and western piedmont adjacent to the range front. Some slopes were built by deposition and others were carved by planation, depending upon the process necessary to achieve the steady- state slope. Two episodes of tectonism disrupted the slope systems. After each episode, stability pre vailed long enough for new steady-state slopes to form at the same angle as the previous equilibrium slopes. In the southern piedmont, tectonism included northward (rangeward) tilt ing of the piedmont which produced segmented alluvial fans. Each fan possesses three seg ments, with the youngest segment at the fan head. Each new segment was built out over the head of the previous segment by deposition until the steady-state slope was achieved. After this, the new slope was time independent until dis rupted by tectonism. Rangeward tilting of the piedmont resulted in slight uplift of the lower piedmont slopes which caused the cutting of terraces that are correlative with the fan seg ments. In the western piedmont, the steady- state slopes were produced by the cutting' of pediment terraces across the piedmont. Three terraces were formed, correlative with the three fan segments and lower piedmont terraces of the southern piedmont. Slight eastward (rangeward) tilting of the piedmont west of Red Ridge may have occurred at this time. A third episode of tectonism resulted in the truncation of the pediment terraces of the western piedmont along a fault which now has 10 m of surface displacement. The zone of faulting broadened to the south and east, and fault scarps are preserved in alluvium of the southern piedmont. Slight eastward tilting of the range and piedmont areas may have occurred at this time and during the Holocene. New pedi ment terraces and fan segments were not initiated by this phase of tectonism. Landscape stability ensued and lasted sufficient ly long for smooth surfaces to form over most valley floor and piedmont slopes. Dunes were developed at this time. On the east side of the range, the Colorado River was entrenched, caus ing truncation of the piedmont. Incision within the mountain block and the removal of valley fill followed landscape stability. Nonpaired fill-strath terraces were cut within the mountain block. In the piedmont, debris removed from the valleys was at first transported across preexisting slopes, and later moved along broad new channels cut into the fans. Deposition occurred between fans, and at the toes of fans, and debris was trans ported out of the piedmont through new channels cut across the lower slopes. Removal of debris from within the range continues today. Incision was produced by an increase in discharge. Dif ferential fluvial erosion of fine-grained sedi ments in the eastern piedmont to produce the present configuration of the terrace terminus most likely occurred at this time, 9* During the Flandrian transgression, a sea cliff was cut into the southeastern portion of the piedmont and across the mouth of Laguna Salada, Transgression extended an unknown but probably short distance up the Colorado River, 10, Depositional regression by the Colorado River resulted in the building of the modern flood plain, burial of the Flandrian shoreline and accretion of muds on Laguna Salada, 2kS Climatic Interpretation Alluviation within the mountain block, the produc tion and maintenance of steady-state slopes in valleys and on the piedmont, landscape stability and incision and the removal of valley fill from within the range were control led by climate. Furthermore, all of the foregoing are be lieved to have occurred during ¥isconsinan and Holocene time. It is possible to use these observations in combina tion with modern rainfall data and general concepts of Wisconsinan climate to infer the nature of climate change through Wisconsinan time in the study area. The present climate was described earlier, and modern rainfall data were presented (Table i). Total rain fall averages only 4.91 cm (l.93 in) Per year. The period of greatest rainfall occurs in the months of August, September and October, and results from the influx of Pacific tropical storms that occasionally move far enough to the north to affect Sierra del Mayor. Rainfall during these months accounts for about 39 percent of the yearly mean. These violent storms produce more intensive rains than occur at any other time of the year, and generate greater runoff than do the gentler rains of other seasons. It is the author's contention that such tropical storms and the runoff generated by them are the key variables of climate throughout most of the Wisconsinan Stage. 249 Tropical cyclones are a mechanism for the northward transfer of heat (Adam, 1975), and are an important element of modern climate. Because typhoons require ocean surface temperatures of 26°C or greater for their production (Palmen, 19^-8; Riehl, 195^+) $ Adam (1975) speculated that during glacial ages when ocean temperatures were reduced, tropical storms might have been far less common than today, or may not have occurred. Other Wisconsinan changes in precipitation are im possible to accurately assess with the present data. Com puter modelling, however, suggests that 18,000 years B.P. the Wisconsinan glacial maximum, the July climate was sub stantially cooler and drier over unglaciated continental areas than the present July climate (Gates, 1976). The present mean July rainfall at Sierra del Mayor is 2 mm, and the means for May and June are 0.2 mm each. A decrease in this slight summer rainfall could be offset by increases in winter or spring rains. As these rains are gentle, how ever, and infiltration is high, slight increases may have had little effect on the landscape. Alternately, precipitation could have been diminished throughout the year during the glacial maximum, as it has been proposed that arid zones expanded, equatorially and that cool humid zones narrowed during this time (Fair- bridge, 1972). In support of a generally drier glacial maximum, evidence of diminished precipitation in the arctic 250 has been presented by Miller (1976), and evidence of tropical ice age aridity has been reported by many scientists (Damuth and Fairbridge, 1970; Parmenter and Folger, 197^> Van Der Hammen, 1972), Palynalogical data is cited by Van Der Hammen (1972) as evidence that during the last glacial maximum, areas of the Amazon basin under tropical rain forest were limited to a number of isolated refugia. Potter, et. al. (1975) have generated a computer simulation of the possible climatic impact of tropical deforestation. Their results indicate overall global cool ing and a reduction in precipitation. Meridional circulation may have been strong in some regions during the glacial maximum (Lamb and Woodroffe, 1970), perhaps blocking normal westerly circulation. ¥inter precipitation at Sierra del Mayor, which accounts for about 33 percent of the yearly mean, is carried in by winter westerlies. If meridional blocking of the westerlies oc curred, winter precipitation could have been greatly reduced. The sequence of climatically controlled events at Sierra del Mayor can be explained by assuming a Sangamonian climate similar to or slightly warmer than the present climate, that tropical storms were either eliminated or in frequent during Wisconsinan time, and that meridional cir culation prevailed during the glacial maximum. The follow ing scenario is proposed. 231 Runoff' generated by fall and late spring tropical storms during the Sangamonian Interglaciation was suf ficient to remove from the mountain block all the debris generated by weathering. Gradual climatic deterioration during early Ifisconsinan time was accompanied by elimina tion or drastic reduction in the number of tropical storms that reached Sierra del Mayor. Peak discharges were no longer sufficient to remove the debris supplied by weather ing at the existing channel slope, even though the weather ing rate may have been slightly reduced by decreased temperatures. As a result, alluviation occurred within the mountain block, until a slope was achieved that was sufficient to transport the debris supplied to the valley floors under the conditions of decreased runoff. Once achieved, these equilibrium or steady-state slopes were maintained until conditions within the drainage basins were altered. During the glacial maximum, meridional circulation tended to block the incursion of westerly storms, and winter precipitation was drastically reduced. When com bined with a reduction in precipitation owing to decrease or elimination of tropical storms, a total decrease of as much as 70 to 75 percent is conceivable. Precipitation at Sierra del Mayor during the glacial maximum may have been less than 1.25 cm per year. Periods of more than one year with no measurable precipitation occur in today's climatic 252 regime, and during the Wisconsinan maximum periods of several years may have elapsed with no measurable rainfall. Under conditions of such extreme aridity, runoff was insuf ficient to maintain bar and channel morphology on active valley floors and piedmont slopes, and desert pavement formed on these surfaces. Dunes were formed on the lower piedmont slopes at this time. Climatic amelioration began sometime after 18,000 years B,P,, and rapid warming ensued around 11,000 years B,P, The number of tropical storms reaching Sierra del Mayor increased as the climate warmed. The increased run off that accompanied the return of the typhoons allowed debris to be removed from valleys within the range at lower gradients than those possessed by the existing valley floors, and progressive incision was the result. 253 REFERENCES CITED 2 5k REFERENCES CITED Adam, D. P., 1975* The tropical cyclone as a global climatic stabilizing mechanism: Geology, v. 3> p. 623-626. Bailey, T. L. and Jahns, R. H. , 1934, Geology of the Trans verse Range Province, Southern California, in Geology of Southern California: Calif. Div. Mines Bull. 170, ch. 2, p. 83-106. Barnard, F. L., 1968a, Structural geology of the Sierra de los Cucapas, northeastern Baja California, Mexico, and Imperial County, California: Colorado Univ. Ph.D. dissertation, 189 p. ________, 1968b, Tectonic history of part of the Gulf of California margin: Geol. Soc. America Program with Abstracts, 68th Annual Meeting, p. 18. Biehler, S., Kovach, R. L., and Allen, C. R., 1964, Geo physical framework of northern end of Gulf of Cali fornia structural province, in van Andel, Tj. H. and Shor, G. G., Jr., eds., Marine Geology of the Gulf of California: Am. Assoc. Petroleum Geologists Mem. 3> p. 126-143. Bloom, A. L., Broecker, ¥. S., Chappell, J. M. A., Mathews, R. K. and Mesolella, K. J., 1974, Quaternary sea level fluctuations on a tectonic coast: New ^30Th/^34u dates from the Huon Peninsula, New Guinea: Quaternary Research, v. 4, p. 183-203. Bull, ¥• B., 1964, Geomorphology of segmented alluvial fans in western Fresno County, California: U. S. Geol. Survey Prof. Paper 332-E, p. 79-129. Buwalda, J. P. and Stanton, ¥. L., 1930, Geological events in the history of the Indio Hills and Salton Basin, Southern California: Science, v. 71> P* 104-106. 235 Chao, S., 1962, Preliminary discussion on the types of gobi in the northwestern part of Ho-Si Tsou-lang and its reclamation and utilization, in Sand Control of the Academia Sinica (ed.), translated by Joint Publica tion Research Service, U.S.A., p. 189-224. Clements, T., Stone, R. 0., Mann, J. F. and Eyman, J. L., 1957» A study of desert surface conditions: Head quarters Quartermaster Research and Development Com mand, Environmental Protection Research Division, Technical Report, EP 53> 110 p. Cooke, R. U., 1970> Stone pavements in deserts: Ann. Assoc. Am. Geog., v. 60, p. 560-577* Cosby, S. ¥., 1929, Notes on a map of the Laguna Salada basin, Baja California: Geog. Rev., v. 19, p. 613-620. Crowell, J. C., 1962, Displacement along the San Andreas fault, California: Geol. Soc. America Spec. Paper 71> 6l p. Curray, J. R., 1965, Late Quaternary history, continental shelves of the United States, in Wright, H. E., Jr., and Frey, D. G., eds., The Quaternary of the United States: Princeton Univ. Press, Princeton, N. J., P. 723-735. Damuth, J. E. and Fairbridge, R. W., 1970, Equatorial At lantic deep-sea arkosic sands and ice-age aridity in tropical South America: Geol. Soc. America Bull., v. 81, p. 189-206. Dibblee, T. W. , Jr. , 195^+> Geology of the Imperial Valley region, California, in Geology of Southern California: Calif. Div. Mines Bull. 170, v. 1, ch. 2, p. 21-28. Doe, B. R., Hedge, C. E., and White, D. E., 1966, Preliminary investigation of the source of lead and strontium in deep geothermal brines underlying the Salton Sea geothermal area: Econ. Geol., v. 6l, p. 462-483. Downs, T. and White, J. A., 1968, A vertebrate faunule succession in superposed sediments from late Pliocene to middle Pleistocene in California: 23rd Internat. Geol. Cong., Prague, Proc,, v. 10, p. 4l-47* 256 Downs, T. and Woodard, G. D., 1961, Middle Pleistocene extension of the Gulf of California into the Imperial Valley, in Abstracts for 1961: Geol. Soc. America Spec. Paper 69» P. 21. Elders, W. A., Rex, R. W., Meidav, T., Robinson, P. T., and Biehler, S., 1972, Crustal spreading in southern California: Science, v. 178, p. 15-24. Fairbridge, R. W., 1972, Climatology of a glacial cycle: Quaternary Research, v. 2, p. 283-302. Garfunkel, Z., 1973> History of the San Andreas fault as a plate boundary: Geol. Soc. America Bull., v. 84, p. 2035-2042. Gastil, R. G., Phillips, R. R., and Allison, E. C., 1975, Reconnaissance geology of the state of Baja California: Geol. Soc. America Mem. 140, 170 p. ________, 1968, Fault systems in northern Baja California and their relation to the origin of the Gulf of Cali fornia, in Conference on geologic problems of the San Andreas fault system, Stanford, California, 1967, Proc.: Stanford Univ. Pubs. Geol. Sci., v. 11, p. 283-286. Gates, W. L., 1976, Modeling the ice-age climate: Science, v. 191, p. 1138-1144. Gile, L. H., Peterson, F. F., and Grossman, R. B., 1966, Morphological and genetic sequences of carbonate ac cumulation in desert soils; Soil Sci., v. 101, p. 347-360. Harvey, J. A. and LaBorde, R. T., 1968, Geology of the Sierra Pinta, Baja California, Mexico: Program, Geol. Soc. America, Cordilleran Sect., 64th Annual Meeting, p. 68. Hastings, J. R. and Humphrey, R. R., 1969, Climatological data and statistics for Baja California: Univ. of Ariz. Inst, of Atm. Physics Tech. Repts. on Meteorology and Climatology of Arid Regions No. 18, p. 9 and 47. Hastings, J. R. and Turner, R. M., 1965, Seasonal precipi tation regimes in Baja California, Mexico: Geografiska Annaler, v. 47A, no. 4, p. 204-223. 257 Hastings, J, R. and Turner, R. M., 1964, Climatological data for Baja California: Univ. of Ariz. Inst, of Atm. Physics Tech. Repts. on Meteorology and Climatology of Arid Regions No. l4, p. 29. Hely, A. G. and Peck, E. L., 1964, Precipitation, runoff and water loss in the lower Colorado River-Salton Sea area: U. S. Geol. Survey Prof. Paper 486-B, p. 1-16. Henyey, T. L. and Bischoff, J. L., 1973, Tectonic elements of the northern part of the Gulf of California: Geol. Soc. America Bull., v. 84, p. 315-330. Hooke, R. LeB., 1972, Geomorphic evidence for Late- Wisconsin and Holocene tectonic deformation, Death Valley, California: Geol. Soc. America Bull., v. 83, p. 2073-2097. ________, 1968, Steady-state relationships on arid-region alluvial fans in closed basins: Am. Jour. Sci., v. 266, p. 609-629. ________, 1967, Processes on arid-region alluvial fans: Jour. Geol., v. 75, p. 438-460. ________, 1965, Alluvial fans: California Institute of Technology, Ph.D. dissertation, 192 p. Hubbs, C. L. and Miller, R. R., 1948, The Great Basin. Part II, the zoological evidence: Bull. Univ. of Utah, v. 38, no. 20, p. 103-113. Ives, R. L., 1949, Climate of the Sonoran Desert: Ann. Assoc. American Geog., v. 39, P. l60. Jaeger, E. C., 1961, The North American Deserts: Stanford Univ. Press, Stanford, Calif., 308 p. Jordan, D. R. and Richardson, R. L., 1907, Description of a new species of killfish, Lucania browni, from a hot spring in Lower California: Proc. U. S. Nat. Museum, v. 33, P. 319-321. Jurwitz, L. R., 1953, Arizona's two season rainfall pat tern: Weatherwise, v. 6, p. 96. Kniffen, F. B., 1932, Lower California studies, 4, the natural landscape of the Colorado delta: Univ. Calif. Publ. in Geog., v. 5, p. 149-244. 258 Kovach, R. L., Allen, C. R. , and Press, P., 1962, Geo physical investigations in the Colorado delta region: Jour* Geophys* Research, v. 67, p. 2845-2871. Krummenacher, D. , Gastil, R* G., Bushee, J., and Doupont, J., 1975, K-Ar apparent ages, Peninsular Ranges batholith, Southern California and Baja California! Geol. Soc. America Bull., v. 86, p. 76O-768. Ku, T. L., Kimmel, M. A., Easton, W. H., and O’Neil, T. J., 197^* Eustatic sea level 120,000 years ago on Oahu, Hawaii: Science, v. 183, p. 959-962. Lamb, H. Ii. and Woodroffe, A., 1970, Atmospheric circula tion during the last ice age: Quaternary Research, v. 1, p. 29-58. Larson, R. L., Menard, H. W., and Smith, S. M., 1968, Gulf of California: a result of ocean-floor spreading and transform faulting: Science, v. l6l, p. 781-784. ________, 1972, Bathymetry, magnetic anomalies, and plate tectonic history of the mouth of the Gulf of Cali fornia: Geol. Soc. America Bull., v. 83, p. 33^5- 3360. Leopold, L. B., Wolman, M. G., and Miller, J. P., 1964, Fluvial processes in geomorphology: V. H. Freeman and Co., San Francisco, 522 p. Lowdermilk, ¥. C. and Sundling, H. L., 1950, Erosion pave ment, its formation and significance: Trans. Am. Geophys. Union, v. 31, P* 96-100. MacDougal, D. T., 1907, Desert basins of the Colorado delta: Bull. Am. Geog. Soc., v. 39* p. 705-729* McKee, E. D. , 1939* Some types of bedding in the Colorado River delta: Jour. Geol., v. 47, p. 64-81. Mabbutt, J. A., 1965, Stone distribution in a stony tableland soil: Austr. Jour. Soil Res., v. 3* P* 131-142. Meckel, L. D., 1975, Holocene sand bodies in the Colorado delta area, northern Gulf of California, in Broussard, M. L., ed., Deltas, models for exploration: Houston Geological Society, p. 239-265. 259 Merriain, R. H. , 1972, Evidence for 200 miles of right lateral displacement on faults of the San Andreas zone in southwest Sonora, Mexico: Geol. Soc. America Abs. with Programs (Cordilleran Section), v. 4, no. 3> P. 198. Merriam, R. H. and Bandy, 0. L., 1965, Source of upper Cenozoic sediments in Colorado delta region: Jour. Sed. Petrology, v. 35* p# 911-916. Metzger, D. G., Loeltz, 0. J., and Irelna, B., 1973* Geo hydrology of the Parker-Blythe-Cibola area, Arizona and California: U. S. Geol. Survey Prof. Paper 486-G, P. 1-130. Miller, G. H., 1976, Anomalous local glacier activity, Baffin Island, Canada: paleoclimatic implications: Geology, v. 4, p. 502-504. Miller, R. R. , 1943 * The status of Cyprinodon raacularius and Cyprinodon nevadensis, two desert fishes of western North America: Occ. Pap. Mus. Zool. , Univ. Mich., v. 473, P. 1-25. Moore, D. G., 1973> Plate-edge deformation and crustal growth, Gulf of California structural province: Geol. Soc. America Bull. , v. 84, p. 1883-1906. Moore, D. G. and Buffington, E. C., 1968, Transform fault ing and growth of the Gulf of California since late Pliocene: Science, v. l6l, p. 1238-1241. Muffler, L. J. and Doe, B. R., 1968, Composition and mean age of detritus of the Colorado River delta in the Salton Trough, southeastern California: Jour. Sed. Petrology, v. 38, p. 384-399# Olmsted, F. H., Loeltz, 0. J., and Irelan, B., 1973, Geo hydrology of the Yuma area, Arizona and California: U. S. Geol. Survey Prof. Paper 486-H, p. 1-227# Orcutt, C. R., 1890, The Colorado Desert: Ann. Rept. Calif. State Mineralogist, v. 10, p. 899-919# ________, 1891, A visit to Lake Maquata: West American Scientist, v. 7, P# 158-164. Palmen, E., 1948, On the formation and structure of tropical hurricanes: Geophysica, v. 3* P# 26-38. 260 Parraenter, C. , and Folger, D. W. , 197^» Eolian biogenic detritus in deep sea sediments; a possible index of equatorial ice age aridity: Science, v. 185, p. 695- 697. Popenoe, W. P., 1954> Mesozoic formations and faunas, southern California and northern Baja California, in Geology of Southern California: Calif. Div. Mines Bull. 170, v. 1, ch. 3, p. 15-21. Potter, G. L. , Ellsaesser, Ii. W. , MacCraclcen, M. C. , and Luther, F. M., 1975> Possible climatic impact of tropical deforestation: Nature, v. 258, p. 697-698. Rice, A., 1976, Insolation warmed over: Geology, v. 4, p. 6l-62. Riehl, H. E., 1954, Tropical meteorology: McGraw-Hill Book Co., New York, 392 p. Roden, G. I., 1964, Oceanographic aspects of Gulf of California, in van Andel, Tj. H. and Shor, G. G., Jr., eds., Marine Geology of the Gulf of California: Am. Assoc. Petroleum Geologists Mem. 3> p. 30-58. Sharon, D., 1962, On the nature of hamadas in Israel: Zeit. fur Geomorph., v. 6, p. 129-147. Springer, M. E., Desert pavement and vesicular layer of some desert soils in the desert of the Lahontan Basin, Nevada: Proc. Soil Sci. Soc. Am., v. 22, p. 63-66. Stanley, G. M., 1963, Prehistoric lakes in Salton Sea basin: Geol. Soc. America Spec. Paper 73> P# 249-250. Stone, R. 0., Carter, L. D., and Yonder Haar, S. P., 1973> Geomorphic analysis of orbital photography of the northern Gulf of California: Zeit. fur Geomorph., Supp. Bd. 18, p. 156-174. Sykes, G., 1937* The Colorado delta: Am. Geog. Soc. Spec. Pub. 19, 193 P. Symmons, P. M. and Hemming, C. F., 1968, A note on wind- stable stone-mantles in the southern Sahara: Geog. Jour., v. 134, p. 60-64. Tarbet, L. A., 1951> Imperial Valley: Am. Assoc. Petroleum Geologists Bull., v. 35* p. 250-253* 261 Tarbet, L. A. and Holman, W. H. , 1944, Stratigraphy and micropaleontology of the west side of Imperial Valley, California: Am, Assoc, Petroleum Geologists Bull,, v. 28, p. 1781-1782. Thomas, R. G., 1963, The Late Pleistocene 150 foot fresh water beach line of the Salton Sea area: Bull. So. Calif. Acad. Sci., Pt. 1, v. 62, p. 9-17* Thompson, R. ¥., 1968, Tidal flat sedimentation on the Colorado River delta, northwestern Gulf of California: Geol. Soc. America Mem. 107, 133 p. van Andel, Tj. H., 1964, Recent marine sediments of the Gulf of California, in van Andel, Tj. H. and Shor, G. G., Jr., eds., Marine Geology of the Gulf of Cali fornia: Am. Assoc. Petroleum Geologists Mem. 3* P* 216-310. Van Der Haramen, T., 1972, Changes in vegetation and climate in the Amazon Basin and surrounding areas during the Pleistocene: Geol. Mijnb., v. 51, p. 641-643• Walker, T. R. and Thompson, R. W., 1968, Late Quaternary geology of the San Felipe area, Baja California, Mexico: Jour. Geol., v. j6, p. 479-485. Woodard, G. D., 1974, Redefinition of Cenozoic stratigraphic column in Split Mountain Gorge, Imperial Valley, Cali fornia: Am. Assoc. Petroleum Geologists Bull., v. 58, p. 521-539. ________, 1963, Geology and stratigraphy of the Vallecito and Carrizo Counties, Southern California: Univ. of California Dept, of Paleontology Ph.D. dissertation. Woodring, W. P., 1938, Distribution and age of the marine Tertiary deposits of the Colorado Desert: Carnegie Inst. Washington Pub. 4l8, p. 1-25. 262 COLORADO FLOOD c^eb ,o S may qs..’; ' ^ vane «>■* ■ * V; f> 2 1 -l\7 \ \ / k . . s i e W ■ 'ffEL • X . Y S V I . MAP SOUTHERN PORTION SIE RRA MAYOR, BAJA CALIFORNIA, MEXICO G E O L O G IC EXPLANATION I Qalf | 1. 1 Silt, sand and clay of the modern Colorado River flood plain. Poorly consolidated, highly saline in places. Qalp - silt, clay, and minor sand of Laguna Salada. Playa, tide flat, and flood plain deposits. Qalps - slightly higher than remainder of playa surface; puffy, salty, halophyte mounds. Fine to medium dune sand. Dunes stabilized but dune surfaces active in places. Sandy alluvium. Moderately to poorly consolidated- Contains sub-rounded to angular clasts up to 5 cm in diameter of granodiorite and schist. Minor soil development at various horizons. Caliche blebs common, gypsiferous in places. Gravelly alluvium. Moderately to poorly consolidated. Contains angular to sub-rounded clasts up to 50 cm in diameter of granodiorite and schist. Minor soil development on various terraces. Caliche blebs and thin pebble coatings common. Silt, sand, and clay of an ancient Colorado River flood plain. Poorly to moderately consolidated. Limy concre tions common in sands. Silts and clays salty and gypsiferous in places. Gravel, sand, silt, and clay of a piedmont-flood plain complex. Thoroughly silicified in places along Red Ridge Fault and along the range front; well consolidated to poorly consolidated elsewhere. Granodiorite. Predominantly biotite granodiorite, but muscovite and garnet common in western part of area. Generally fine grained, but pegmatitic in places. Highly frac tured; joints spaced from less than 2 cm to as much as 1 m apart, generally less than 30 cm apart. Undifferentiated metamorphics. Biotite schist and minor amphibolite, gneiss, and chlorite schist; very minor guartzite. Highly fractured- weak, can often be broken by hand. Contact Fault, dashed where inferred, dotted where concealed. Ball on downthrown side. Amount and direction of dip indicated where known Crushed zone Fault scarp in alluvium, hachures on downthrown side. Querried where mapped by photo interpretation Graben in alluvium Horst in alluvium Ground breaks observed on 1949 air photos Lineation on ERTS photography or conventional air photos Bedding attitude, with dip in degrees Trend and plunge of minor fold Escarpment in alluvium, dots at base of slope Shoreline Well Mine Federal Highway 2 £00 G Pla.fe 1

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Evaporites and algal mats at Laguna Mormona, Pacific Coast, Baja California, Mexico

PDF

Seismic stratigraphic study of the California Continental Borderland basins: Structure, stratigraphy, and sedimentation

PDF

Tectonics and geochemical exploration for heavy metal deposits in the Southern Gulf of California

PDF

Metamorphism in the Big Maria Mountains, southeastern California

PDF

Estuarine sediment transport and Holocene depositional history, Upper Chesapeake Bay, Maryland

PDF

Quaternary geomorphic surfaces on the northern Perris Block, Riverside County, California: Interrelationship of soils, vegetation, climate and tectonics

PDF

Paleogeography of the Southern California forearc basin in the late Cretaceous and early Paleogene: Evidence from paleomagnetism and carbonate facies analysis

PDF

Late Cenozoic tectonics of the central Mojave Desert, California

PDF

Trends in extent and depth of bioturbation in Great Basin Precambrian-Ordovician strata, California, Nevada and Utah

PDF

A sedimentologic history of Ridge Basin, Transverse Ranges, southern California

PDF

Late Precambrian diabase intrusions in the southern Death Valley region, California: Their petrology, geochemistry, and tectonic significance

PDF

Stratigraphic and sedimentologic analysis of the Monterey Formation: Santa Maria and Pismo basins, California

PDF

Holocene stratigraphy and sedimentary processes in Santa Barbara Basin: Influence of tectonics, ocean circulation, climate and mass movement

PDF

Mesozoic and Cenozoic extensional tectonics of the Halloran and Silurian Hills area, eastern San Bernardino County, California

PDF

Geochemical and sulfur isotope study of Red Sea geothermal systems

PDF

Processes influencing transportation and deposition of sediment on the continental shelf, Southern California

PDF

Structure of Santa Cruz-Catalina Ridge and adjacent areas, Southern California Continental Borderland from reflection and magnetic profiling: Implications for late Cenozoic tectonics of Southern...

PDF

Seismic sequence stratigraphy and structural development of the southern outer portion of the California Continental Borderland

PDF

The rheology of single crystal sodium chloride at high temperatures and low stresses and strains

PDF

Geology of a Tertiary volcanic center, Mopah Range, San Bernardino County, California

Asset Metadata

Creator Carter, Louis David (author)

Core Title Late Quaternary erosional and depositional history of Sierra del Mayor, Baja California, Mexico

Degree Doctor of Philosophy

Degree Program Geological Sciences

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Geomorphology,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c29-342817

Unique identifier UC11219112

Identifier DP28541.pdf (filename),usctheses-c29-342817 (legacy record id)

Legacy Identifier DP28541.pdf

Dmrecord 342817

Document Type Dissertation

Rights Carter, Louis David

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA