Tectonic geomorphology of the zone in the southern Indio Hills, ,

E A KELLER • M S BONKOWSKI I Department of Geological Sciences, University of California, Santa Barbara, California 93106 R. J. KORSCH* Armidale College of Advanced Education, Armidale, New South Wales 2350, Australia R. J. SHLEMON P.O. Box 3066, Newport Beach, California 92663

ABSTRACT INTRODUCTION

Geomorphic investigation of the San Andreas fault zone in the Tectonic geomorphology studies along the San Andreas fault Indio Hills indicates many tectonically produced landforms, includ- in the Indio Hills began in the winter of 1978-1979. Objectives of ing beheaded streams, right-lateral deflected and offset streams, this paper are: (1) to delineate patterns of recent tectonic deforma- sags, shutter ridges, pressure ridges, and fault scarps. Near Biskra tion and possible relations to simple shear; (2) to evaluate evidence Palms, an alluvial fan-pediment complex has an apparent cumula- for paleoseismicity in the area; and (3) to estimate the slip rate for tive offset of about 0.7 km along the Mission Creek fault zone this section of the San Andreas fault. Two areas were mapped in (north branch, San Andreas fault). Many of the tectonic landforms, detail: Pushawalla Canyon and an uplifted, deformed, and offset as well as the fracture pattern that has developed during the Pleisto- pediment-fan complex. cene, are explainable by simple shear or uplift associated with a The Indio Hills trend northwest along the northeast flank of small left bend in the main trace of the Mission Creek fault. The the Coachella Valley, California (Fig. 1). The hills are cut by and ratio of vertical to horizontal displacement in the vicinity of the uplifted along several branches of the San Andreas fault zone, most bend is about 0.04. notably the Banning and Mission Creek faults that intersect near Exposed in Pushawalla Canyon, 5 km northwest of the alluvial Biskra Palms (Fig. 2). fan-pediment complex, are: (1) a sequence of stream terraces, (2) The southern Indio Hills southeast of Thousand Palms folded Plio-Pleistocene fanglomerates,, and (3) an example of Canyon are pri marily composed of the Pleistocene Ocotillo Forma- stream capture following a right lateral deflection or offset of sev- tion (Fig. 2), a deformed sequence of sandstone and fanglomerates eral hundred metres. A left step of the Mission Creek fault in (Dibblee, 1954; Popenoe, 1959). Pushawalla Canyon is a probably cause of folds and sporadic uplift Topography of the Indio Hills is characteristic of arid regions that produced the stream terraces. Possible cause of recent stream (rainfall is 7 to 10 cm/yr) undergoing recent tectonism and fluvial capture are: (1) juxtaposition of Pushawalla Canyon with a relict erosion. The major fluvial landforms are steep canyons and coalesc- canyon moving northwestward along the Mission Creek fault, or ing alluvial fans along the north west-southfeast trending front of the (2) right-lateral deflection or offset of Pushawalla Canyon along the hills. Upland surface and older alluvial fans are covered with fault, with simultaneous headward erosion of a shorter, steep moderate- to well-developed desert pavement and desert varnish. stream flowing toward the Coachella valley. However, the most impressive aspect of the topography is an Estimation of a slip rate and identification of paleoseismicity assemblage of "classic" tectonically produced landforms associated for the San Andreas fault in the Indio Hills is difficult. However, with the San Andreas fault zone. Landforms observed are: fault degree of topographic dissection, formation, and preservation of scarps; beheaded, deflected, or offset drainages; tectonically desert pavement, and relative soil profile development suggest that induced stream capture; sags; shutter ridges; and pressure ridges. the age of the offset fan may be as old as 70,000 yr but that most Paleoseismicity of the San Andreas fault in the Indio Hills is likely it is on the order of 20,000 to 30,000 yr. These age estimates unknown. The main fault traces are shown on Figure 2, and these for the offset fan indicate a minimum slip rate for the San Andreas are being studied in detail (Bonkowski, unpub. data) to gather fault of 10 to 35 mm/yr, with about 23 to 35 mm/yr the most likely. information concerning recent earthquake activity. The main fault The pattern of observed offset drainages is complex, but suggests trace is locally delineated by fan palms (Washingtonia filifera) that that during the past few thousand years creep events or moderately form linear oases along the water-bearing fault zones. Offset drain- large earthquakes have periodically produced several metres of ages at several locations along the Mission Creek fault indicate that right-lateral displacement. creep or moderate magnitude earthquakes have occurred in the past few thousand years. The offset is generally 1 to 4 m but the history •Present address: Geology Department. Victoria University, Welling- is complex, making the data difficult to interpret. ton, New Zealand. Most of the smaller earthquakes (M <3.5) in recent years

Geological Society of America Bulletin, v. 9!!, p. 46-56, 10 figs., 3 tables, January 1982.

46

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/93/1/46/3434348/i0016-7606-93-1-46.pdf by guest on 30 September 2021 TECTONIC GEOMORPHOLOGY OF SAN ANDREAS FAULT ZONE 47

oi o -+- 34°00' Figure 1. Index map, southern Indio Hills. CA

q Desert Hot Springs 0'/}

apparently occurred north of the main trace of the San Andreas fault (Fig. 3), and the area has no known history of great earth- quakes. There have been moderate to large seismic events, including the 1948 Desert Hot Springs earthquake, which had a magnitude of 6.5 (Fig. 3). Therefore, the large active fault system does represent a potential hazard to rapidly urbanizing local areas, including Palm strongly developed, typified by reddish, clay-rich B (argillic) horiz- Springs, Desert Hot Springs, Palm Desert, and Indio. ons, and often by two or more discrete carbonate (calcic) horizons (Buol and Yessilsoy, 1964; Elam, 1974; Gile, 1968; Gile and others, OFFSET PEDIMENT-ALLUVIAL FAN COMPLEX 1965, 1966). Soils of this relative development are usually pre-late Pleistocene in age, and deemed "relict paleosols," for they usually An uplifted, deformed, and offset pediment-alluvial fan com- occur on high-level, remnant surfaces, and were never buried; plex is located southeast of Biskra Palms near the junction of the rather, they attained most of their development during environ- Banning and Mission Creek faults (Figs. 2 and 4). The area contains ments of the past (Ruhe, 1965; Valentine and Dalrymple, 1976). a spectacular assemblage of fault-related landforms, including fault Some soils in the Mojave Desert and adjacent areas have been scarps; beheaded, deflected, and offset streams; shutter ridges; sags; dated radiometrically, by radiocarbon or U-Series (Bischoff and and microtopography (horst and graben?). These are shown dia- others, 1978; Ku and others, 1979). Others have been dated rela- grammatically on Figure 5. tively, generally by association to Pleistocene climatic change, as The minimum cumulative offset of the fan Qf2 is about 0.7 km recored by the number, depth, and morphology of pedogenic car- (Fig. 5, x-x^. The northern part of the modern fan (Qfi) is appar- bonate horizons (Arkley, 1963; Gile and others, 1965, 1966; Morri- ently offset several hundred metres, suggesting that the right lateral son, 1978; Morrison and Frye, 1965; Nettleton and others, 1975; slip continued into the Holocene. Shlemon, 1978b). Soil ages at one locality, however, cannot usually The surface [Qf2 and Qf2(?), Figs. 2 and 5] is designated a be directly applied to another, owing to differences in parent mate- pediment-fan complex because the upper segments are a true ero- rial lithology and grain size, precipitation, slope, and other basic sion surface with only a thin veneer of deposits a metre or so thick soil-forming factors. Nevertheless, nonsodic, well-drained soils in overlying the Ocotillo Formation. The upper pediment and lower fan the Mojave Desert less than about 15,000 yr old are often typified surfaces are tentatively correlated by relative soil development, by disseminated carbonates in the upper part of the profile, but lack topographic dissection, and formation of desert pavement and other distinctive pedogenic horizons. In contrast, relict profiles on desert varnish. surfaces more than about 70,000 or 80,000 yr old usually have an Soil (pedogenic profiles) are useful indicators to estimate the argillic horizon (Bt), two or more carbonate horizons (Bca; Cca), age of desert fan and pediment surfaces (Bull, 1974, and unpub. and often a stone-free, vesicular (A2 or "Av") horizon (Bischoff and data; Gile, 1975, 1977; Gile and Hawley, 1966; Nettleton and others, others, 1978; Bull, unpub. data; Peterson, 1980; Shlemon, 1978a; 1975; Shlemon, 1978a). Many desert soils are moderately to Springer, 1958). Accordingly, such differences in relative profile

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/93/1/46/3434348/i0016-7606-93-1-46.pdf by guest on 30 September 2021 48 KELLER AND OTHERS

development make it possible to date, at least approximately, fans Pit no. 1; Fig. 5) is characterized by incipient development (A-C and pediments in the study area. Thus, several soil profiles were profile), and only a very thin, vesicular (Av) horizon. The desert measured and described from pits excavated on various surfaces pavement is moderately developed. However, the fan surface is now (Fig. 5). Tables 1, 2, and 3 present three detailed profiles. The inactive, having been offset in a right-lateral direction by the San profile terminology and field classification are standard, following Andreas fault. The age of this fan, based mainly on the weakly that of the U.S. Soil Conservation Service (Soil Survey Staff, 195!, developed soil profile, is deemed Holocene. 1975). The proximal segments of the pediment-fan complex contain All soils on the pediment-fan complex (with one exception excellent examples of tectonically produced landforms. The surface near the toe of the offset fan) are relict paleosols (Typic Haplar- has been modified only slightly by fluvial, eolian, and mass wasting gids). The profiles have well-developed argillic horizons (Bt) with processes. Thus, it provides a rare opportunity to observe and study superimposed, younger pedogenic carbonates (Bca¡ and Btca; landforms associated with strike-slip faulting that have undergone

Tables 1 and 2). The soil on the Qf2 surface (Pit no. 4; Fig. 5; Table I) minimal modification by fluvial processes. However, not all the has profile characteristics similar to that on the Qf2(?) surface (Pit topography on the upper surface (Qf2(?); Fig. 5) is tectonically gen- no. 6; Fig. 5; Table 2). Most diagnostic are moderately developed, erated; relict channels and mudflow deposits also exist. reddish argillic horizons extending to a depth of at least 40 cm. The major tectonic landforms that exist on the upper surface

Modern pedogenic carbonates are restricted to the upper part of the (Qf2(?) are visible on low sun-angle aerial photography (Fig. 4) and argillic horizons, reflecting the present shallow depth of wetting are shown diagrammatically on Figure 5. The surface has been (IIBca and Btca horizons; Tables 1 and 2). These soils and the uplifted about 30 m, most likely in response to convergence asso- surfaces upon which they are forming are thus judged to have been ciated with the small left bend of the Mission Creek fault (Figs. 2 subjected to at least one interval of "soil pluviality." Soils with and 5). The sags and microtopography (horst and graben?) are comparable development elsewhere in the Mojave Desert are possible extensional features produced in accordance with a simple thought to be less than about 70,000 yr old, and generally in the shear model (Fig. 6) slightly modified after Wilcox and others range of about 20,000 to 30,000 yr old (Bischoff and others, 1978; (1973); Crowell (1974); Sylvester and Smith (1976); Dibblee (1977). Bull, 1974 and unpub. data; Ku and others, 1979; Nettleton and The microtopographic forms may be bounded by small normal others, 1975; Shlemon, 1978a). faults, but examination of the features in natural cuts has not dem- A soil excavated in the offset portion of the modern fan (Qfi, onstrated that fault traces definitely exist. However, due to the very

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/93/1/46/3434348/i0016-7606-93-1-46.pdf by guest on 30 September 2021 TECTONIC GEOMORPHOLOGY OF SAN ANDREAS FAULT ZONE 49

San Andreas Fault Zone | Qol | Quaternary stream gravel Southern Indio Hills | QT | Quaternary terrace QT, -QT2

| Qf11 Quaternary fan

jlQfz] Quaternary fan-pediment complex

|'Po'.| Ocotillo Formation

anticline (and plunge) ^ thrust (teeth on Figure2.(Continued). • syncline (and plunge) r upper plate) y boundary fault, dashed where ^ strike and dip fgL sv^-' inferred, dotted where 10 of beds Q burried U-up, D-down

coarse texture of the gravels, definite fracture zones are less likely to colluvium overlying a calcic horizon. However, modern carbonate be recognized. Certainly, the scarps are younger than the surface for sediment accumulated at a depth of only 15 cm. This suggests that there is little or no development of desert varnish on many of them. the linear basin periodically receives fine sediment from the adja- We tested the hypothesis that the scarps might be relic stream cent scarps that border the possible microtopographic horsts. channels by digging a pit in the floor of one of the linear basins (Pit Microtopography is quite variable, but the basins are several metres no. 5; Fig. 5). The upper 40 to 45 cm consists of fine sand and silty wide, with scarps about 1 m high. The simple shear model combined with uplift associated with the 116°30' 116° 10' 34° 34° small left bend of the fault can explain the origin of many tectonic features of the offset fan. It also explains the orientation of the x Xx X • X X * ^ * Desert Hot Figure 3. Selected earthquake epicenters and magnitudes for Springs I earthquakes in the Indio Hills area. All events 1970 to 1975 and 1948 Desert Hot Springs earthquake, M = 6.5. Data from Califor- nia Institute of Technology, Program run by Zhao-Liang Zhang (personal commun.).

1932 - present M >5.0 N Palm Springs 5 km 1970-75 X - M >3.5 < 4.5 X = M > 2.5 < 3.5 X = M > 1.5 < 2.5 33° 5km 45' x = M< I . 5 —I

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/93/1/46/3434348/i0016-7606-93-1-46.pdf by guest on 30 September 2021 50 KELLER AND OTHERS

broad syncline in the Ocotillo Formation between Pushawalla Canyon and Thousand Palms Canyon (Fig. 2). However while the simple shear model is attractive, it is only a first approximation to the solution of a complex problem. For example, many of the folds in the Indio Hills are parallel to the Mission creek fault rather than trend at an oblique angle as would be predicted by a simple shear model. We also tested the hypothesis that simple Shear influences the development of fractures by measuring the orientation of 127 nearly vertical bed-rock fractures and comparing their distribution to expected patterns of main trace shears, synthetic shears, antithetic shears, and P shears. Figure 7 A shows the expected orientations of shear fractures associated with right-lateral strike slip faulting; and Figure 7B shows the pattern of fractures observed in the Pleistocene bed-rock of the Indio Hills. These data tend to support the simple shear model and suggest that fracture orientation can be influenced over a relatively long time (about 1,000,000 yr). The Indio Hills is a good location to test the simple shear model because the rocks are young and therefore recent structure is well exposed and not imprinted on older deformation patterns. On the other hand, frac- tures are difficult to recognize in the Ocotillo Formation and little information concerning length and sense of displacement could be obtained from fieldwork. Therefore, the relationship of the frac- tures to the simple shear model (Figs. 7A and 7B) is tentative. The pattern of shears (Fig. 7B) does suggest that the main trace and R Figure 4. Low sun-angle photograph of offset pediment- (synthetic) shears are most clearly delineated, resembling the pat- alluvial fan complex. Notice the beheaded streams on the fan, the tern of residual structure in the shear zone of the Dasht-e Bayaz fault prominent fault scarp and large sag on the upper surface. Addi- zone following the 1968 earthquake of magnitude 7.2 and some tional landforms are shown on Figure 5. The darker surface of the shear box experiments (Tchalenko, 1970, his Fig. 11). fan with well-developed desert pavement and desert varnish is in marked contrast to the light-colored modern alluvium. Photograph PUSHAWALLA CANYON courtesy of Woodward-Clyde Consultants. The walls of the Pushawalla Canyon show two traces of the Mission Creek fault (north branch of the San Andreas fault). These

Figure 5. Sketch map showing the assemblage of landforms produced by tec- tonic processes. Sites where soil profiles were described are shown by the numbers 1 through 8.

fQ* ) (

0 100 m TUP approximate scale • 2 location of soil pit

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/93/1/46/3434348/i0016-7606-93-1-46.pdf by guest on 30 September 2021 TECTONIC GEOMORPHOLOGY OF SAN ANDREAS FAULT ZONE 51

Figure 6. Simple shear model to ex- plain some of the landforms produced by Contraction«] I features £3 Extensional features right-slip along the San Andreas fault. -e— FOLDS NORMAL FAULTS \\ k PUSH UPS PULL APARTS \ì V SQUEEZE UPS £ HORSTSGRABENS

traces are related to a left step in the fault system (Fig. 8). The In the northwest canyon wall, southwest of the Fi trace of the southern trace (Fi; Fig. 8) is a wide gouge zone with discrete shears. fault, the angular unconformity between the lower and upper Oco- Bedding in the upper Ocotillo Formation, normally gently dipping, tillo Formation is well exposed. is steepened within the zone. Locally, small sets of normal faults Several terraces occur in Pushawalla Canyon and its main showing displacements of about 50 mm occur between two major tributary; two are mapped on Figure 2. The overbank and channel shears. The northern trace (F2; Fig. 8) occurs also as a wide gouge deposits of the pre-capture stream in Pushawalla Canyon are the zone, and within the zone, beds are steepened to almost vertical. most extensive and herein are named "Qti." Qt2 is a prominent Farther to the south in the southeast wall of Pushawalla Canyon, terrace along the main channel and tributary. The absolute ages of a wide (as much as 100 m) fault zone strikes obliquely to the main the terraces are presently unknown, but the Pleistocene age of the traces of the San Andreas fault (N20°W compared with N50° W for Ocotillo Formation which the stream has dissected gives an older the San Andreas fault). This zone is characterized by several dis- limit. Less extensive desert varnish on the lower terraces, compared crete fault traces which have two predominant dips of about 45° with the upper terraces, may indicate a high rate of downcutting and 65° south. Bedding generally is truncated by the faults. How- owing to both stream capture and/or local uplift. Inspection of the ever, within the zone, where the faults steepen, the dips of the beds soils developed on the terrace materials indicates very little profile are also steepened. development. Therefore, we infer they are Holocene in age, that the

TABLE 1. SOIL PROFILE DESCRIPTION, INDIO HILLS AREA, PIT NO. 4

Horizon Depth (cm) Description*

Av 0-4 Light gray (10YR 7/2) to very pale brown (10YR 7/4) when moist, silt; massive structure; soft, very friable, nonsticky and nonplastic; clear smooth boundary. 11 Bea 1 4-13 Strong brown (7.5YR 5/6) to brown (7.5YR 4/4) when moist, gravelly sandy loam; massive structure; soft, very friable, slightly sticky and nonplastic; strongly effervescent; few to common, thin carbonate rinds on bases of pebbles; gradual wavy boundary.

IIBca2 13-26 Similar to IIBca, with common, thin carbonate rinds on bases of pebbles, and few, medium soft lime masses; gradual wavy boundary.

IIB2t 26 - 36 Brown (7.5YR 4/4) to yellowish-red (5YR 4/6) when moist, gravelly fine sandy loam; weak, fine to medium subangular structure; hard, friable, slightly sticky and nonplastic; few, thin clay films lining tubular and interstitial pores, and colloidal staining of mineral grains; gradual wavy boundary. IIB-.-C 36 - 40+ Yellowish-brown (10YR 5/6) to dark yellowish-brown (10YR 4/6) when moist, gravelly and fine sandy loam grading to sandy loam near base; massive structure; soft, friable, nonsticky and nonplastic; very few, thin clay films lining tubular pores; in gradual diffuse boundary; base of pit.

Note: location is shown in Figure 5. •Soil description from field measurements. Colors based on Munsell notation. Profile terminology follows that of Soil Survey Staff (1951).

TABLE 2. SOIL PROFILE DESCRIPTION, INDIO HILLS AREA, PIT NO. 6

Horizon Depth (cm) Description*

Av 0 - 3 Light gray (10YR 7/2) to very pale brown (10YR 7/4) when moist, silt; massive structure; soft, very friable, nonsticky and nonplastic; abrupt smooth boundary. Btca 3 - 15 Strong brown (7.5YR 5/6) to brown (7.5YR 4/4) when moist, very fine sandy loam; weak, medium subangular blocky structure; soft, friable; slightly sticky and nonplastic; very few, thin clay films lining tubular pores and staining mineral grains; violently effervescent; common, fine irregular soft lime in masses (stage II carbonate development); gradual wavy boundary. 15 - 41+ Brown (7.5YR 4/4) to yellowish-red (5YR 4/6) when moist, fine sandy loam; weak, medium, subangular structure grading B2, to massive near base; soft, friable; slightly sticky and nonplastic; very few, thin colloidal clay films on mineral grains, and linging tubular and interstitial pores; few, fine, rounded soft lime masses in lower 8 cm; base of pit.

Note: location is shown in Figure 5. •Soil description from field measurements. Colors based on Munsell notation. Profile terminology follows that of Soil Survey Staff (1951).

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/93/1/46/3434348/i0016-7606-93-1-46.pdf by guest on 30 September 2021 52 KELLER AND OTHERS

TABLE 3. SOIL PROFILE DESCRIPTION 2, BURIED PALEOSOL, MASSEY QUARRY, WEST WALL, INDIO, CALIFORNIA

Horizon Depth (cm) Description*

®21cab 0- 13 Light brown (7.5YR 6/4) to brown (7.5YR 4/4) when moist, heavy sandy loam; weak, fine granular to subangular blocky structure; slightly friable, slightly sticky and nonplastic; common, thin clay films lining tubular pores; colloidal staining on mineral grains; violently effervescent; common, medium segregated lime filaments; gradual irregular boundary. l*22b 13 - 27 Yellowish red (5YR 5/6) to reddish brown (5YR 5/4) when moist, sandy loam; weak, fine to granular subangular blocky structure; slightly hard, friable, slightly sticky and nonplastic; fe w, thin clay films lining tubular pores; colloidal staining on mineral grains; slightly effervescent; few, fine segregated lime filaments and lime threads; gradual irregular boundary. ®3b 27 - 53 Strong brown (7.5YR 5/6) to brown (7.5YR 4/4) when moist, light sandy loam; massive to weak, fine granular to sub- angular blocky structure; slightly hard, firm nonsticky and nonplastic; very few, thin colloidal clay strains; diffuse irregular boundary. IIClb 53 - 76 Light yellowish-brown (10YR 6/4) to yellowish-brown (10YR 5/4) when moist, loamy sand; massive structure; granular, loose when dry and moist; nonsticky and slightly plastic; noncalcareous; horizon contains common mafic (schist) clasts near base; abrupt wavy boundary grading laterally to clear wavy.

IIC2cab 76 - 140 Very pale brown (10YR 7/3) to light yellowish-brown (10YR 6/4) when moist, sand; single-grained; loose when dry and moist, nonsticky and nonplastic; violently effervescent; 25-35 percent pebbles and cobbles; many, medium lime seams; large, rounded soft lime masses; calcareous rinds to 2 mm on pebble bases in lower part of horizon (stage III carbonate development); gradual irregular boundary. "ICjb 140 - 146 Very pale brown (10YR 7/3) to light yellowish brown (10YR 6/4) when moist, coarse sand; single-grained; loose when dry and moist; nonsticky and nonplastic; noncalcareous; gradual irregular boundary. 146 - 155 White (10YR 8/2) to very pale brown (10YR 8/3) when moist, pebbly silty loam; single-grained; loose when dry and moist; nonsticky and nonplastic; violently effervescent; horizon is laterally discontinuous; medium disseminated lime; gradual wavy boundary.

VC5 155 - 160+ Pebbly coarse sand with silt lenses; base of cut.

Note: location is shown in Figure 5. •See text for general description of enclosing sediment; soil description from field measurement, colors based on Munsell notation; profile terminology follows that of Soil Survey Staff (1951).

stream capture was a Holocene event, and that the deflection or between Hidden Palms and Pushawalla Canyon, was deflected offset of Pushawalla Canyon was a late Pleistocene event. and/or offset about 1 km prior to capture (Figs. 8 and 9). We The excellent exposures of the semi consolidated Ocotillo For- hypothesize that the downstream portion of the present drainage, mation in Pushawalla Canyon result from recent incision by the downstream of the capture to the desert floor, is a relict canyon present stream following capture. The abandoned channel, seen which was transported northwestward along the fault. Alterna-

Figure 7. (A) Geometry of folds and faults to a right-slip faul t. Modified after Sylvester and Smith, 1976. (B) Rose diagram of fault and shear traces between offset alluvial fan and Thousand Palmis Canyon Road. MT = MAIN TRACE. (R) = SYNTHETIC SHEAR. (P) = P SHEAR. (R) = ANTITHETIC SHEAR. 127 field measurements by Brunton method.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/93/1/46/3434348/i0016-7606-93-1-46.pdf by guest on 30 September 2021 Figure 8. Stream profiles and geologic section of the Pushawalla Canyon area. See text for explanation.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/93/1/46/3434348/i0016-7606-93-1-46.pdf by guest on 30 September 2021 54 KELLER AND OTHERS

Figure 9. Topographic map showing the left step of the San Andreas fault and abandoned chan- mp? nel of Pushawalla Canyon, Myoma 7.5-min Quadrangle, California.

tively, however, the capture could have been induced by rapid little evidence for the F2 fault trace was observed, presumably headward erosion of a small steep channel trending southwest to because it dies out between Pushawalla Canyon and Tributary 3. the Coachella Valley. Certainly, the probability of such a capture In Pushawalla Canyon, the uplift has been over a larger area increases as the amount of right-lateral deflection increases. than just the area between the two fault traces. This uplift may have Most streams in the study area (Fig.. 8) exhibit normal concave produced the folding observed in the Pushawalla Canyon walls profiles (for example, Creek 1, Creek 2, Tributary 1, Tributary 2, (Fig. 8). Here the upper Ocotillo Formation has been folded into and Thousand Palms Canyon). The two streams which do not fol- broad gentle anticlines and synclines with the two major fault traces low this pattern are Pushawalla Canyon and Tributary 3, which occurring in the southwest limbs of the anticlines. The convex pro- have convex profiles in their central parts. This change in profile files suggest the possibility of very recent deformation, because the from concave to convex possibly can be attributed to compression, stream should rapidly erode to a concave profile. resulting from the left step in the right-lateral San Andreas fault An additonal example of stream capture is also shown on the system in the vicinity of Pushawalla Canyon. The convex nature is stream profiles (Fig. 8, Creeks 1 and 2). The abandoned channel in more pronounced in Pushawalla Canyon than in Tributary 3, where the canyons to the northwest of Pushawalla Canyon and between

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/93/1/46/3434348/i0016-7606-93-1-46.pdf by guest on 30 September 2021 TECTONIC GEOMORPHOLOGY OF SAN ANDREAS FAULT ZONE 55

350 presented by Crowel (1952 and 1975) and Ehlig, Ehlert, and Crowe (1975), using the time scale of Berggren and Van Couvering (1974), are shown in Figure 10. The rate of 20 to 30 mm/yr is an average for -303 ± 5Km the past 12 m.y. If the assumption is made that from 12 m.y. ago to about 4 m.y. ago the principal strand of the system was the San Gabriel fault (Crowell, 1975), then a slip rate of closer to 60 mm/yr is suggested for the past 4 m.y. This rate is consistent with the hypothesis that about 4 m.y. ago the San Andreas fault began to play a transform role in the opening of the Gulf of California (Moore and Buffington, 1968; Crowell, 1975). The rate of spreading of the Gulf, assumed to be about 60 mm/yr (Moore and Buffington, 25 i 3 mm/yr 1968), is transformed to strike-slip deformation along the San Andreas fault system. Our best estimate of slip (23 to 35 mm/yr) at the offset alluvial fan near Biskra Palms in the southern Indio Hills accounts for only about one-half of the Pliocene-Pleistocene slip rate of the San Andreas system; thus, another centimetre or so per year of slip may be taken up on faults other than the Mission Creek A- Ehlig, Ehlert and Crowe,1975 fault. However, it is emphasized that the estimated slip rate for the Crowell, 1975 Indio Hills is not definitive. Further, the average slip rate provides

Crowell, 1952 little information concerning paleoseismicity, that is, the number and magnitude of Quaternary earthquakes. The offset fan near Biskra Palms provides information con- cerning the ratio of vertical to horizontal slip on the San Andreas 15 20 25 30 fault. The fan has been uplifted about 30 m while moving a min- imum of about 0.7 km right laterally. Thus, the ratio of recent M.Y. vertical to horizontal slip at this site is ~0.04. Figure 10. Possible displacement rates for the southern San The modern fan, well exposed at the gravel pit southeast of Andreas fault. Compiled by T. Davis (personal commun.). See text Biskra Palms (Fig. 5), also provides information concerning the for explanation. recent history of the San Andreas fault. The pit exposes late Pleis- tocene and Holocene sediments about 30 m thick. The stratigraphic Creeks 1 and 2 exhibit normal concave stream profiles and appar- history is that of a "stacked" alluvial fan; that is, we presume that ently have not been associated with the uplift caused by right-lateral the deposits in the lower section have been transported along the movement at the left step in the San Andreas fault. fault. Thus, at times in the past the wedge of fan material was alternately receiving sediment from upstream canyons or moving DEFORMATION RATES AND PALEOSEISMICITY between canyons. During times when sediment was not being added, the surface was stable and soil profiles formed. We thus The slip rate and paleoseismicity for the San Andreas fault in would expect to observe wedges of sediment separated by buried the Indio Hills have not been absolutely determined. However, sev- soils. eral lines of investigation provide insight: (1) estimation of the age The gravel pit exposes at least four buried paleosols as well as of the offset fan; (2) investigation of buried soils in a fan exposed in a deeper relict-weathering front. The most prominent buried soil the gravel pit adjacent to the offset fan; and (3) evaluation of small occurs at a depth of about 15 m and is clearly visible throughout the offset drainages along recent active strands of the San Andreas pit. The profile is incompletely preserved; the organic and part of fault. the "B" horizon have been eroded. Investigation of the desert pavement and soil development on This buried paleosol (Table 3) is very strongly developed the offset alluvial fan (discussed above) suggests an age of as much (Paleargid); a profile nearly 200 cm thick is preserved. Its reddish as 70,000 yr, but more likely is 20,000 to 30,000 yr old, and thus is color (5YR 5/6), blocky structure, clay films (cutans) and stage III associated with the most recent major pluvial epoch (National calcic horizon (IIC2cab) attest to either development over a long Academy of Sciences, 1977). These estimates imply an avg slip rate time, or formation in an environment more conducive to profile on the San Andreas fault of ~ 10 to 35 mm/yr, with a rate of about formation than at present. At a minimum, this soil is at least 23 to 35 mm/yr, corresponding to the 20,000 to 30,000 yr. age 100,000 yr old, although conceivably several times that age (Bull, considered the best estimate. We also estimated a longer-term slip 1974, and unpub. data; Nettleton and others, 1975; Shlemon, rate from an apparent 27 km offset of fold axes in the Palm Springs 1978a). Formation. Assuming an age of 1 million yr for the formation, then the slip rate is about 30 mm/yr. However, it is emphasized that the CONCLUSIONS case for the offset is speculative and that the age of the folds and their specific relation to the San Andreas fault zone is not well 1. The simple shear model is sufficient to explain the orienta- known. tion of many bed-rock fractures as well as the remarkable assem- Estimates of slip rates for the southern San Andreas fault sys- blage of landforms produced by strike-slip faulting in the Indio tem compiled by T. Davis (1980, personal commun.) from data Hills.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/93/1/46/3434348/i0016-7606-93-1-46.pdf by guest on 30 September 2021 56 KELLER AND OTHERS

2. Vertical deformation is observed at several locations where Gile, L. H., 1968, Morphology of an argillic horizon in desert soils of the fault trace bends or steps left. At one location, the ratio of southern New Mexico: Soil Science, v. 106, p. 6-15. vertical to horizontal displacement is 0.04. 1975, Holocene soils and soil-geomorphic relations in an arid region of southern New Mexico: Quaternary Research, v. 5, no. 3, p. 321-360. 3. The San Andreas fault in the Indio Hills has offset many 1977, Holocene soils, and soil-geomorphic relations in a semi-arid re- drainages 1 to 4 m,suggesting that there has been creep or moderate gion of southern New Mexico: Quaternary Research, v. 7, no. 1, earthquakes on a time scale of a few thousand years. p. 112-132. 4. A definitive slip rate for the San Andreas fault in the Indio Gile, L. H., and Hawley, J. W., 1966, Periodic sedimentation and soil for- mation on an alluvial fan piedmont in southern New Mexico: Soil Hills is not available. An estimated range for the slip rate of 10 to 35 Science Society of America Proceedings, v. 53, p. 261-268. mm/yr is based on a 0.7 km cumulative offset of an alluvial fan Gile, L. H„ Peterson, F. F„ and Grossman, R. B„ 1965, The K-Horizon—A whose age is estimated by soil profile development to as much as master horizon of CaCOj accumulation: Soil Science, v. 99, p. 74-82. 70,000 yr, but which most likely is about 20,000 to 30,000 yr. The 1966, Morphological and genetic sequences of carbonate accumulation latter age provides an estimated slip rate of 23 to 35 mm/yr. in desert soils: Soil Science, v. 101, p. 347-360. Ku, T. L„ Bull, W. B., Freeman, S. T„ and Knauss, K. G., 1979, Th23°-U234 dating of pedogenic carbonates in gravelly desert soils of Vidal Valley, ACKNOWLEDGMENTS southeastern California: Geological Society of America Bulletin, pt. 1, v. 90, no. 11, p. 1063-1073. Reviews of the manuscript and suggestions for improvement: Moore, D. G., arid Buffington, D. C., 1968, Transform faulting and growth of the Gulf of California since the late Pliocene: Science, v. 161, by W. B. Bull and members of the tectonic geomorphology class of p. 1238-1241. 1979 at the University of Arizona, W. A. Elders and J. E. Kahle, are: Morrison, R. B., 1978, Quaternary soil stratigraphy: Concepts, methods, appreciated. B. Baca, M. Clark, T. Dibblee, Jr., W. Hart, P. Link, and problems, in Mahaney, W. C., ed., Quaternary soils: Geological A. MacDonald, L. Minck, V. Ramirez, D. Reseigh, and T. Rock- Abstracts, University of East Anglia, Norwich, England, p. 77-108. well are acknowledged for field assistance and preliminary interpre - Morrison, R. B., and Frye, J. C., 1965, Correlation of middle and late Qua- ternary successions of the Lake Lahontan, Lake Bonneville, Rocky tation developed during a field mapping exercise as part of a Mountain (Wasatch Range), southern Great Plains and eastern tectonic geomorphology class at the University of California, Santa Midwest areas: Nevada Bureau of Mines Report 9, 45 p. Barbara, winter 1979. Special recognition and appreciation are National Academy of Sciences, 1977, Climate, climatic change and water extended to George L. Meyer, who brought the offset alluvial fan to supply: Studies in Geophysics, 132 p. our attention in 1978. Nettleton, W. D., Witty, J. E., Nelson, R. E., and Hawley, J. W„ 1975, Genesis of argillic horizons in soils of desert areas of the southwestern United States: Soil Science Society of America Proceedings, v. 39, REFERENCES CITED no. 5, p. 919-926. Peterson, F. F., 1980, Holocene desert soil formation under sodium salt influence in a playa-margin environment: Quaternary Research, v. 13, Arkley, R. J., 1963, Calculation of carbonate and water movement from soil no. 2, p. 172-186. climatic data: Soil Science, v. 96, no. 4, p. 239-248. Berggren, W. A., and Van Couvering, J. A., 1974, The late Neogene: Pa- Popenoe, F. W. , 1959, Geology of the southeastern portion of the Indio Hills, Riverside County, California [M.A. thesis]: Los Angeles, Cali- laeogeography, Palaeoclimatology, Palaeoecology, v. 16, p. 1-216. fornia, University of California at Los Angeles, 153 p. Bischoff, J. L., Childers, W. M., and Shlemon, R. J., 1978, Comments on the Pleistocene age assignment and associations of a human burial from the Ruhe, R. V., 1965, Quaternary paleopedology, in Wright, H. E., and Frey, Yuha Desert, California: A rebuttal: American Antiquity, v. 23, no. 4, D. G., eds., The Quaternary of the United States: Princeton, New p. 747-749. Jersey, Princeton University Press, p. 755-764. Bull, W. B., 1974, Geomorphic tectonic analysis of the Vidal region, in Shlemon, R. J., 1978a, Quaternary soil-geomorphic relationships, south- Woodward-McNeill and Associates, Vidal Nuclear Gen. Station, Units eastern Mojave Desert, California and Arizona, in Mahaney, W. C., 1 and 2. Appendix 2.5B (Geology and Seismology), Southern Cali- ed., Quaternary soils: Geological Abstracts, University East Anglia, fornia Co., Rosemead, California, 66 p. Norwich, England, p. 187-207. Buol, S. W., and Yessilsoy, M. S., 1964, A genesis study of the Mojave sandy 1978b, Buried calcic paleosols near Riverside, California, in Geologic loam profile: Soil Science Society of America Proceedings, v. 28, guidebook to the Santa Ana River basin, southern California: South p. 254-256. Coast Geological Society (Irvine, California), Annual Field Trip Guide- book, p. 32-40. Crowell, J. C., 1952, Probable large lateral displacement on the San Gabriel fault, Southern California: American Association of Petroleum Geolo- Soil Survey Staff, 1951, Soil survey manual: U.S. Department of Agricul- gists Bulletin, v. 36, p. 2026-2035. ture Soil Conservation Service, Agriculture Handbook 18, 503 p. 1974, Origin of Late Cenozoic basins in Southern California, in Tec- 1975, Soil taxonomy: U.S. Department of Agriculture, Soil Conserva- tonics and sedimentation: Society of Economic Paleontologists and tion Service, Agriculture Handbook 436, 754 p. Mineralogists Special Publication 22, p. 190-204. Springer, M. E., 1958, Desert pavement and vesicular layer of some desert soils in the desert of the Lahontan Basin, Nevada: Soil Science Society 1975, The San Andreas fault in Southern California, in Crowell, J. C., of America Proceedings, v. 22, p. 63-66. ed. San Andreas fault in Southern California: California Division of Sylvester, A. G., and Smith, R. R., 1976, Tectonic transpression and base- Mines and Geology Special Report 118, p. 7-27. ment-controlled deformation in San Andreas fault zone, Salton Dibblee, T. W., Jr., 1954, Geology of the Imperial Valley region, California: Trough, California: American Association of Petroleum Geologists California Division of Mines and Geology Bulletin 170, p. 21-28. Bulletin, v. 60, p. 2012-2018. 1977, Strike-slip tectonics of the San Andreas fault and its role in Techalenko, J. S., 1970, Similarities between shear zones of different magni- Cenozoic Basin evolvement, in Late Mesozoic and Cenozoic sedimen- tudes: Geological Society of America Bulletin, v. 81, p. 1625-1634. tation and tectonics in California: Sar. Joaquin Geological Society Valentine, K.W.G., and Dalrymple, J. B., 1976, Quaternary buried soils: A Short Course, p. 26-38. critical review: Quaternary Research, v. 6, p. 209-222. Ehlig, P. L., Ehlert, K. W., and Crowe, B. M., 1975, Offset of the Upper Wilcox, R. E., Harding, T. P., and Seely, D. R., 1973, Basic wrench tecton- Miocene Caliente and Mint Canyon Formations along the San Gabriel ics: American Association of Petroleum Geologists Bulletin, v. 57, and San Andreas faults, in Crowell, J. C., ed., San Andreas fault in p. 74-96. Southern California: California Division of Mines and Geology Special Report 118, p. 83-92. MANUSCRIPT RECEIVED BY THE SOCIETY JUNE 5, 1980 Elam, N. E., 1974, Soil survey—Palo Verde area, California: Washington, REVISED MANUSCRIPT RECEIVED APRIL 6, 1981 D.C., U.S. Department of Agriculture Soil Conservation Service, 37 p. MANUSCRIPT ACCEPTED APRIL 10, 1981

Printed in U.S.A.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/93/1/46/3434348/i0016-7606-93-1-46.pdf by guest on 30 September 2021