Processes of oscillatory basin ﬁ lling and excavation in a ...pangea.stanford.edu/.../HilleyandStrecker_GSAB2005.pdf · rocks. In the vicinity of the Toro basin, uplifted basement

For permission to copy, contact [email protected]© 2005 Geological Society of America 887

ABSTRACT

Intramontane basins may act as important sediment storage areas, serve as recorders of the history of deformation, record syntectonic deposition, and document the evolution of climatic conditions during deposition. We document the timing, cyclicity, and processes that led to the fi lling and reexcavation of the intramontane Quebrada del Toro basin in NW Argentina. Geomorphic and geologic observations indicate that the basin was fi lled with sediment that has been subsequently excavated at least two times in the last ~8 m.y. The last fi lling and excavation cycle occurred within the last 0.98 m.y. and has led to the deposition and removal of ~61.4 km3 of material from the basin, leading to a basin-wide averaged minimum denudation rate of 0.16 mm/yr. Aggradation within the basin took place due to channel steepening of the downstream fl uvial system that connects the intramontane basin to the foreland. This por-tion of the fl uvial system is actively incising through an uplifting bedrock zone. We use observations within the Toro to test a quasi-physically based model of channel aggrada-tion behind a rising base level that rises due to downstream channel steepening. Our work shows that the bedrock incision rate constant required to reproduce conditions observed within the Toro basin is consistent with val-ues measured independently in similar rock types. Therefore, in intramontane basins that experience similar processes of fi lling and evacuation, this model may be used to assess the relative importance of tectonic rock uplift,

bedrock resistance to fl uvial incision, and climate in determining the geomorphic and sedimentologic history of these basins.

Keywords: intramontane basin, contrac-tional orogen, bedrock incision, internal drainage, sedimentology.

INTRODUCTION

Fault-bounded intramontane basins are an integral part of thrust belts and often form behind actively growing topography. They are therefore important recorders of deforma-tion, erosion, and syntectonic deposition. Due to the dynamic nature of such environments, sedimentary units contained in these basins have abundant unconformities, growth strata (e.g., DeCelles et al., 1991; Suppe et al., 1992; Jordan et al., 1993), and pronounced variation in clast size and sediment type (e.g., DeCelles and Giles, 1996). In addition, the basin sedi-ments may record important climatic changes related to orographic barrier uplift (Kleinert and Strecker, 2001; Starck and Anzótegui, 2001).

Documentation of the processes, timing, episodicity, and cyclicity of fi lling and reex-cavation of these basins is often hampered by a lack of radiometrically datable strata and the pervasiveness of coarse clastic units not ame-nable to magnetostratigraphic dating. In some intramontane basins located in the hinterland of volcanic arcs, however, the fortuitous combina-tion of preserved landforms and abundant inter-calated volcanic ashes may resolve the timing of fi lling and reexcavation, tectonic activity, and geomorphic evolution necessary to understand the dynamics of these depozones. In addition, these observations may be used to evaluate quasi–process-based models of the evolution of developing intramontane basins (e.g., Sobel et al., 2003). Both observations and theoretical predictions may subsequently be used to under-

stand the general principles of intramontane basin evolution in other regions.

In this study, we document the processes of basin fi lling and reexcavation within the Quebrada del Toro basin (hereafter referred to as the Toro basin) in northwestern Argentina at ~24°S. This basin is located between the eastern Puna border and the southern Eastern Cordillera (Fig. 1) and is hydrologically con-nected to the foreland by the Río Toro through a narrow, fault-bounded bedrock gorge. Within the basin, the timing and processes of fi lling behind a rising topographic barrier are well recorded by structural, stratigraphic, and geo-morphic relationships.

We fi rst constrain the timing and processes of the steepening and defeat of the range-travers-ing fl uvial system. Secondly, we document the fi lling of basins formed behind the fragmented network, the subsequent recapture of the basin by headward erosion, and fi nally the reexca-vation of the basin fi ll. Thirdly, we use these observations to test the model of channel defeat and basin fi lling proposed by Sobel et al. (2003) for structurally similar basins in the adjacent arid Puna Plateau. Our study shows that this simple model may be used in other basins to understand the relative importance of basin geometry, climate, uplift, incision processes, and the resistance of bedrock to fl uvial incision in the establishment of internal drainage.

METHODS

We document hydrologic isolation and reintegration of the Toro basin by mapping of geologic units, compilation of radiometric ages, Quaternary surfaces, and active structures. In addition, we use sedimentologic observations to constrain the depositional environment, source, and composition of the basin-fi ll deposits, and Digital Elevation Model (DEM) analysis of the topography to characterize the geometry

Processes of oscillatory basin fi lling and excavation in a tectonically active orogen: Quebrada del Toro Basin, NW Argentina

George E. Hilley†

Department of Earth and Atmospheric Sciences, University of California, 377 McCone Hall, Berkeley, California 94707, USA

Manfred R. Strecker‡

Institut für Geowissenschaften, Universität Potsdam, Postfach 601553, 14415 Potsdam, Germany

GSA Bulletin; July/August 2005; v. 117; no. 7/8; p. 887–901; doi: 10.1130/B25602.1; 10 fi gures; 2 tables.

†Present address: Department of Geological and Environmental Sciences, Braun Hall, 450 Serra Mall, Building 320, Stanford, California 94305, USA.

‡Corresponding author e-mail: [email protected].

HILLEY and STRECKER

888 Geological Society of America Bulletin, July/August 2005

of the basin. Relationships between faults, the deposition of basin-fi ll units, and geomorphic surfaces were mapped on ~1:25,000 scale aerial photography and rectifi ed CORONA and LANDSAT TM satellite images and compiled into a Geographic Information System (GIS) for map generalization and presentation. Paleo-current directions within the sediments were inferred from clast imbrications collected at 19 sites throughout the basin. At each site, ~50 clast imbrications were measured and corrected for any tilting and folding of the sedimentary units. Clast compositions of sedimentary units at 21 sites within the basin were used to constrain temporal changes in sediment sources. To deter-mine the elevations of the different geomorphic surfaces and the volume of sediment removed by river incision, we created 1:25,000 scale con-tour maps. These contour maps were combined with mapped channels in the area and converted to 30-m-resolution DEMs using the Topogrid algorithm (Hutchinson, 1989) in the Arc/INFO software package. Total volume removal during basin excavation was calculated by reconstruct-ing and extrapolating the highest terrace surface across the basin to determine the level to which the basin was formerly fi lled and subtracting the present-day topography from this surface. The DEM was also used in conjunction with the geo-logic maps to construct along-channel profi les.

GEOLOGIC OBSERVATIONS OF FILLING AND EXCAVATION

Structural, Sedimentologic, Geomorphic, and Climatic Setting

The Toro basin is one of a series of reverse-fault-bounded basins between the arid Puna Plateau to the west and the humid foreland to the east (Figs. 2 and 3). Plio-Pleistocene contraction and uplift east of the Puna have contributed to the compartmentalization of the foreland into intramontane basins and interven-ing ranges such as the Toro basin and its sur-rounding highlands (Strecker et al., 1989; Mar-rett et al., 1994; Kleinert and Strecker, 2001). Deformation is accommodated within bivergent reverse-fault-bounded mountain blocks that uplift Precambrian through Paleozoic basement rocks. In the vicinity of the Toro basin, uplifted basement lithologies include weakly metamor-phosed Eocambrian fl ysch units (Puncoviscana Formation), Proterozoic granite, and Paleozoic quartzite (Mesón and Santa Victoria groups), locally covered by Upper Cretaceous to Eocene sandstones and carbonates (mainly Yacoraite Formation of the Salta Group) (Reyes and Salf-ity, 1973). Over the Paleozoic and Cretaceous units lies a series of late Miocene basin sedi-

ments that were part of a continuous foreland basin until ca. 6.4 Ma (magnetostratigraphic study area shown as PS in Fig. 3; Viramonte et al., 1994; Reynolds et al., 2000). However, clast provenance in the upper portion of this sequence indicates that the Sierra Pasha to the east (Fig. 4) was being unroofed prior to this time (R. Alonso, 2004, personal commun.). The infl ux of clasts derived from Sierra Pasha to basins farther east commenced sometime between 8.73 Ma and 7.5 Ma (Viramonte et al., 1994) and constitutes a thickness between 500 and 1300 m. Along the eastern margin of the ranges between this site and the Toro basin (Fig. 3B), the contact between the Puncovis-cana Formation and the overlying Mesón group is currently between 3000–3500 m above the location of the east-verging fault responsible for the uplift of the units (Mikuz, 2002). In addition, a series of uplifted bedrock terraces at the eastern range front along the Río Toro (loca-tions shown in Fig. 4; features shown in Fig. 5) indicate that this uplift is ongoing. Therefore, faulting along the eastern margin of the ranges bounding the Toro region has produced a total of between 3500 and 4800 m of rock uplift in the last 8.73–7.5 m.y., leading to a long-term aver-age rock-uplift rate with respect to the footwall of 0.40–0.64 mm/yr.

The funnel-shaped Toro basin contains a rich and well-exposed section of syntectonic

basin-fi ll units that have been deformed by repeated faulting along the margins and within the interior of the basin (Fig. 3A; Marrett et al., 1994; Marrett and Allmendinger, 1990; Marrett and Strecker, 2000). In general, the shape of the Toro basin has resulted from a reorienta-tion of maximum shortening in the region from WNW prior to the Quaternary to ENE during the Quaternary (Marrett and Allmendinger, 1990; Marrett et al., 1994; Marrett and Strecker, 2000). The western and northwestern margin of the basin is bordered by the east-verging Solá and San Bernardo faults, respectively, while the west-verging Gólgota fault marks the eastern boundary between the basin and the uplifting range to the west (Fig. 3A; Schwab and Schäfer, 1976; Marrett and Strecker, 2000). The middle Miocene foreland strata within the basin belong to the Agujas Conglomerate that contains gra-nitic and volcanic units from the ranges bound-ing the Puna. This indicates that tectonic activity on the eastern edge of the Puna and in the proto-Eastern Cordillera (Marrett and Strecker, 2000) was ongoing during this time.

The present channel elevation of the Río Toro is between 2000 and 2650 m; however, remnants of once contiguous geomorphic surfaces exist up to 600 m above the channel. The surround-ing ranges have elevations in excess of 5000 m, and in the case of Sierra Pasha (Fig. 3A), were repeatedly glaciated during the Pleistocene.

Figure 1. Tectonic map of the central Andes (modifi ed after Isacks, 1988). The internally drained Altiplano-Puna plateau is bordered by a series of intramontane basins within the Cordillera Oriental and Sierras Pampeans. Box shows the location of Figure 2, Star denotes Quebrada del Toro.

PROCESSES OF OSCILLATORY BASIN FILLING AND EXCAVATION

Geological Society of America Bulletin, July/August 2005 889

Downstream from the Toro basin and the sedi-mentary fi ll units, the Río Toro crosses the south-ern part of the Gólgota Fault and enters a narrow gorge that exposes basement rocks (Figs. 3A and 3B). The gorge terminates at the fault-bounded mountain front of the Sierra Pasha Sur, while the river fl ows eastward into the intramontane Lerma Valley (Fig. 3A). Thus, the upper reaches of the Río Toro traverse highly erodible Tertiary and Quaternary sediments while the downstream portions incise metamorphic basement rocks exposed south of the Gólgota Fault.

The meridianal orientation of the Eastern Cordillera concentrates moisture along its eastern fl ank, creating a steep precipitation gradient between the humid foreland and the arid Toro basin (Fig. 2B). Mean annual pre-cipitation within the Lerma Valley (Fig. 3A) is ~1200 mm/yr, while the leeward Toro basin receives an average of 260 mm/yr (WMO, 1975). Similar to other basins along the Puna, the outlet of the Toro basin allows penetration of moisture into the gorge and ultimately the Toro basin.

Basin Filling and Excavation

The lowest 200 m of the Plio-Pleistocene basin-fi ll section comprises sandstones and intercalated conglomerates that laterally grade into fern fossil-bearing lacustrine deposits in the southern part of the basin near Estación Solá (Fig. 4). This unit unconformably overlies the Miocene deposits that are older than 8 Ma (Marrett and Strecker, 2000). The presence of these fossils in the lower section of the Pliocene unit suggests that annual precipitation was

Figure 2. (A) Shaded relief topography of NW Argentina. Heavy dashed line shows the boundary between the internally drained Puna-Alti-plano Plateau and the externally draining wedge top and foreland basins that drain to the east. Box shows extent of Figure 3. (B) Colored and contoured mean annual precipitation distribution. Headwaters of basins typically lie within arid portions of the landscape, while down-stream outlets traverse narrow gorges that coincide with high gradients in mean annual precipitation. Based on data from WMO (1975) and Bianchi and Yañez (1992); modifi ed after Haselton et al. (2002), topography from U.S. Geological Survey (GTOPO30).

HILLEY and STRECKER


signifi cantly greater than now (R. Alonso, 2001, personal commun.). These fl uvio-lacustrine units are conformably overlain by coarse con-glomerates (hereafter referred to as the Alfarcito Conglomerate) that contain a 4.17 Ma tuff in the lower part of the conglomeratic section (Mar-rett and Strecker, 2000). The Alfarcito Con-glomerate and the underlying units are locally deformed by faulting along the basin margins and within the basin. Growth strata within the conglomerate that contain a volcanic tuff dated 0.98 Ma (Marrett et al., 1994) indicated that deposition took place during deformation.

Undated Quaternary conglomerates at least 700 m thick are separated from the underlying Alfarcito Conglomerate by an angular uncon-formity. Pronounced relief (>100 m) along the unconformity suggests that partial removal of the Alfarcito Conglomerate preceded the deposition of these younger Quaternary units. Currently, the highest exposures of the Qua-ternary conglomerate unit are ~700 m above the current base level, and the upper section of these outcrops forms a series of deeply dissected erosional surface remnants. Therefore, after the deposition of this unit (post–0.98 Ma), the Río Toro has cut through its entire thickness and

exposed the Alfarcito conglomerate, Miocene strata and basement rocks. Miocene volcanic rocks and overlying Quaternary boulder con-glomerates are exposed along the axis of a northward-plunging anticline (Fig. 4). The ero-sional resistance of these units preserves their high topography, whereas areas to the east and west are deeply incised (Figs. 3 and 4).

We plot the paleofl ow directions within the lower and upper Alfarcito Formation and the Quaternary Conglomerate in Figures 6A, 6B, and 6C, respectively. The lower Alfarcito Con-glomerate (<8–4.17 Ma) is a clast-supported, medium-bedded conglomerate with subrounded to rounded clasts. The clasts are generally well imbricated, and lack of a well-defi ned and repeating fi ning-upward sequence suggests that this section was deposited during west to east fl ow in a gravel-dominated, low-sinuosity fl uvial environment (Fig. 6A). Clast litholo-gies are dominated by rocks of the extensively exposed Puncoviscana Formation. A relatively minor component of quartzite currently found in the eastern basin-bounding ranges suggests that there was no signifi cant exposure of this lithol-ogy to the east (Fig. 6A). The upper portion of the Alfarcito conglomerate (deposited between

4.17 and 0.98 Ma; Fig. 6B) consists of a clast-supported medium-pebble conglomerate con-taining subrounded to subangular clasts. Sandy interbeds ~40 cm thick are intercalated with the conglomeratic beds. Clast imbrications are well developed and indicate a transport direction dominantly from north to south in a gravelly bedload-dominated environment (Fig. 6B). Source lithologies are dominated by the Pun-coviscana Formation, whereas the abundance of rock types typical for the Puna margin at this latitude, such as porphyritic volcanics and green volcanic breccias, is similar to that of the lower Alfarcito Conglomerate.

Finally, the timing of deposition of the Quaternary Conglomerate is bracketed by the 0.98 Ma 40Ar/39Ar age in the underlying deformed Alfarcito Conglomerate (Marrett et al., 1994) and a 30 ka lacustrine unit depos-ited within a terrace inset into the Quaternary Conglomerate (Trauth and Strecker, 1999). The Quaternary Conglomerate is a matrix-supported, poorly sorted and bedded deposit. Clasts are angular to subangular, and average clast size generally increases with proximity to the range front. Fluvial reworking and occa-sional fl uvial deposition allowed us to measure

Figure 3. (A) Geologic units and structures of the Toro basin. Sierra Pasha consists of Cambro-Ordovician quartzites and Precambrian weakly metamorphosed Puncoviscana Formation (adapted from Marrett and Strecker, 2000, and references therein). (B) Shaded relief topography of Toro basin (SRTM data, U.S. Geological Survey) showing location of photographs in Figure 5 (letters with arrows), the locations of Figures 4 and 6, and the location of the paleomagnetic stratigraphic site (PS) reported in Viramonte et al. (1994) and Reynolds et al. (2000).



Figure 4. Landsat TM image of the Toro basin showing the major geographic features, locations, and station numbers where paleofl ow and provenance measurements were collected and general geomorphology of the basin. The constriction of the Río Toro at the southern portion of this image corresponds to the location of the exposed Puncoviscana Formation in the Sierra Pasha Sur.

HILLEY and STRECKER


transport directions within this unit (Y stations in Fig. 6C). Clast transport directions are from ENE to WSW, and the major component of the source lithology consists of quartzites of the Mesòn Group sourced from the Sierra Pasha. The Quaternary Conglomerate probably repre-sents a series of proximal to medial debris-fl ow deposits intercalated with fl uvial units deposited in front of the uplifting ranges.

The relationship between the fault-bounded Puncoviscana Formation within the gorge, the post–0.98 Ma Quaternary Conglomerate, and geomorphic surfaces within the Toro basin and along the gorge south of it indicates that deposi-tion of this conglomerate resulted from steep-ening of the channel within the gorge and the fi lling of the upstream Toro basin. First, uplifted strath terraces at the downstream outlet of the Río Toro indicate that the Sierra Pasha Sur has

been undergoing uplift during the Quaternary (Fig. 7A). Second, geomorphic surfaces within the Toro basin have not formed in bedrock, but rather have been cut into Plio-Pleistocene fi ll units (Fig. 7C). Therefore, basement ranges bounding the Toro basin have been and continue to undergo relative rock uplift. In contrast, while it may be part of regional warping related to uplift of the adjacent Puna, the Toro basin has not experienced this degree of uplift, result-

ing in aggradation as the downstream fl uvial system steepens (e.g., Humphrey and Konrad, 2000). In addition, the currently observed chan-nel gradients within the uplifting bedrock zone are steeper than those north of the Gólgota Fault (Fig. 7B). The highest gravel-covered erosional surface remnants of the Quaternary Conglomerate comprise gently inclined sur-faces that appear to buttress the rising bedrock barrier. These geometric relationships, the

Figure 5. (A) Upstream view of uplifted strath terraces along the outlet of the Río Toro as it debouches from the Sierra Pasha Sur. Arrows show the location of one set of paired terraces and a higher, unpaired terrace. (B) View of the transition between fi ll and strath terraces in the vicinity of the constriction of the Río Toro noted in Figure 4. Two downward-pointing arrows highlight the location of geomorphic surfaces underlain by a moderately thick fi ll unit. However, the unconformity between the mid-Miocene Agujas conglomerate and overlying fi ll units indicates that the Río Toro has cut into the uplifting basement wedge. The downstream surface (downward-pointing arrow at far left) is underlain by rocks of the Puncoviscana Formation. (C) View of geomorphic surfaces upstream of the Sierra Pasha Sur. Virtually all geomorphic surfaces are cut into the sedimentary fi ll of the Alfarcito and Quaternary conglomerates. Arrows highlight locations of several geomorphic surfaces beveled into the fi ll units. For locations and orientations of photographs, see Figure 3B.

Figure 6. Provenance and paleofl ow in late Miocene–Quaternary sediments exposed within the Toro basin. In all panels, rose diagrams indicate direction of paleofl ow. Provenance at each station is shown as a colored bar, the fractions of which represent the relative contri-bution of a particular lithology. Panels show the provenance and paleofl ow within the (A) lower Alfarcito conglomerate, (B) upper Alfarcito conglomerate, and (C) Quaternary Con-glomerate. (D) Rose diagrams showing combined paleofl ow measurements from each unit.



HILLEY and STRECKER


consistent local clast source of the Quatermary Conglomerate, and paleofl ow directions suggest that the basin was probably either closed or on the threshold of being hydrologically isolated during the late stages of the deposition of the Quaternary Conglomerate.

Erosional removal of the Plio-Pleistocene conglomerates created a series of at least seven inset fl uvial terraces within the basin and noncorrelative strath terraces within the Río Toro gorge (Fig. 7) and thus was episodic. Figure 8 shows the spatial distribution of mate-rial removed since the abandonment of the highest erosional surface. The total volume of sediment removed since deposition of the Quaternary conglomerate is 61.4 km3. This erosion has taken place since 0.98 Ma, lead-ing to a minimum basin-wide averaged ero-sion rate of 0.16 mm/yr. However, because the minimum age of the last basin-wide aggradation episode is 0.98 Ma and erosion is consequently younger, this rate represents an absolute minimum average rate and is probably much higher.

Relationships between the locus of incision, the geologic structures, and different lithologies can be seen in Figure 8. First, the lowermost sections of the Río Toro trunk stream (Fig. 8, label A) have begun to incise through the underlying unconformity with the Tertiary units and has thus reoccupied the base level prior to deposition of the Plio-Pleistocene conglomer-ates. Second, maximum erosion in the basin is concentrated within E-W trending tributaries and areas to the east (label B) of the N-S–ori-ented anticline (Fig. 8). The tributaries that have incised through the Miocene volcaniclastic rocks within the anticline are constricted. Where the downcutting of these streams has reached areas to the east of the anticline, erosional removal of unconsolidated or weakly consolidated units is effi cient (Fig. 8, label C). In these areas, rela-tively planar erosional surfaces are graded to the base level of each tributary.

Synthesis of Geologic and Geomorphic Data

The regional tectonic and stratigraphic relationships in the greater Toro basin area record the transition from a foreland basin to an intramontane basin, where sediments sub-sequently folded after 6.4 Ma (Viramonte et al., 1994; Reynolds et al., 2000), when uplift within the Sierra Pasha apparently created suffi cient topography to starve the foreland of Puna-derived lithologies (Fig. 9). The uplift of the Sierra Pasha was accompanied by deforma-tion at least partially accommodated along the proto-Solá fault on the west and subsequent erosion of the Tertiary section in the Toro

basin. This was followed by deposition of the lacustrine units and the lower Alfarcito conglomerate. During this time, uplift within the Sierra Pasha was apparently insuffi cient to create signifi cant topography to have blocked moisture transport and eastward paleofl ow as recorded within the lower Alfarcito unit. The fact that topography existed within the Sierra Pasha (Viramonte et al., 1994; Reynolds et al., 2000) and that there are only eastward and southward paleofl ow directions in the

sediments suggests that the position of the axial trunk stream was east of the exposed lower Alfarcito outcrops (Fig. 9).

By the time of the deposition of the upper Alfarcito conglomerate (<4.17 Ma), defor-mation had been renewed within the basin, resulting in folding and syntectonic deposition of conglomerates. The character of these sedi-ments suggests that they were deposited in a broken foreland setting in which uplift of the range to the east disrupted the W-E fl owing

Figure 7. (A) Schematic geologic cross section along the Río Toro. Upper line delimiting gray area shows the smoothed maximum elevation envelope along the river, while the lower line corresponds to the relative elevation of the Río Toro. (B) Relative elevation and geom-etry of geomorphic surfaces (circles, squares, triangles, and diamonds). Within the uplifting bedrock gorge, only noncorrelative strath terraces (fi lled circles) are observed, whereas upstream of the Solá Fault, a series of abandoned geomorphic surfaces record downcutting of the Río Toro and excavation of the basin. Points show location of spot elevation measure-ments from DEM, while lines show geometry of the surface and are dashed where inferred.



channels and forced fl ow parallel to the strike of the uplifting range. Despite a rising down-stream base level, the lack of lacustrine inter-calations and southward-directed paleofl ow indicators suggests that the basin probably remained externally drained. After 0.98 Ma, thrusting along the Solá Fault and the basin margins deformed the entire Alfarcito sec-tion. Associated with these events was erosion recorded by the angular unconformity between the deformed Alfarcito conglomerate and over-lying units (Fig. 9).

The excavation of the Alfarcito conglomer-ate provided space for renewed aggradation of coarse conglomerates sometime after 0.98 Ma. The composition and paleofl ow indicators of these deposits suggest renewed relative uplift or climate changes affecting the surrounding ranges or a combination thereof. However,

following deposition, the conglomerates were incised as documented by terraces between 600 m and ~30 m above the present base level (Fig. 9).

In summary, these observations document cyclic fi lling and reexcavation of the Toro basin that has recurred over millions of years. While the base of the Toro basin has risen over its history, most of the rock uplift in the area is accommodated within the surrounding ranges. In response to this uplift, the portions of the Río Toro channel traversing these ranges have apparently steepened, resulting in aggradation in the upstream Toro basin. Furthermore, the spatial coincidence of the high topography and the steep precipitation gradients currently observed suggests that surface uplift within this range has created steep precipitation gra-dients with ensuing aridifi cation.

MODELING BASIN FILLING AND THE ESTABLISHMENT OF INTERNAL DRAINAGE IN THE TORO BASIN

Model Formulation

Observations from the Toro basin indicate that basin fi lling and perhaps eventual hydro-logic isolation from the foreland resulted from the inability of aggradation within the basin to keep pace with uplift within the southern Sierra Pascha. This situation is analogous to the establishment of internal drainage in the Puna Plateau proposed by Sobel et al. (2003). In their formulation, uplift within a rising, bedrock-cored zone induces simultaneous aggradation behind the rising barrier and fl uvial incision into bedrock within the uplifting area (Fig. 10). They found that the establishment of internal drainage is determined by a threshold that depends on the ratio of the rock uplift within the bedrock-cored zone to the square root of the product of the transport constant of the channel and bedrock incision constant in the following manner:

U

KDc

=1

2Son+ 12 πWb( )

12ka

m 1− hm( )12

Wb +Wu( )1−hm−Wb

1− hm[ ]12. (1)

Variables and units in equation 1 are listed in Table 1. In a qualitative sense, as the rock-uplift rate increases relative to the transport effi ciency and incision capability of channels (left-hand side of equation 1), internal drainage is increas-ingly favored. The threshold between internal and external drainage is defi ned by a unique value of U KDc, and so values of U KDcgreater than and less than this will result in the hydrologic isolation of these basins and preser-vation of external drainage, respectively.

The Toro basin presents a unique opportunity to test this model because (1) the process of aggradation due to downstream channel steep-ening is directly observed within this basin; (2) geologic observations and DEM calculations allow constraints to be placed on most variables in the formulation; and (3) the Toro basin has fl uctuated between large magnitude aggradation and incision over the past ~8 m.y. Therefore, the basin probably lies close to its threshold of hydrologic isolation, and small changes in uplift rate, rock erodibility conditions, and/or precipi-tation may lead to downstream channel steep-ening, upstream aggradation, and hydrologic isolation. Under these conditions, the U KDc ratio appropriate for the Toro basin should be captured by equation 1.

Unlike the model by Sobel et al. (2003), the Toro basin has had repeated cycles of fi lling

Figure 8. Distribution of sediment removed from Toro basin since abandonment of the highest geomorphic surface formed after the deposition of the post–0.98 Ma conglomerate. This distribution was calculated by fi tting a surface to the remnants of the topographically highest geomorphic surface and subtracting the current topography from this surface. Sur-face remnants are highlighted by areas of low exhumation (dark blue). The total volume of material removed from the basin within >0.98 m.y. is 61.4 km3. Labels A, B, and C refer to locations discussed in the text.

HILLEY and STRECKER


and reexcavation, and so the model must be modifi ed to account for the effects of preexist-ing downstream topography and basin-incision events. We consider only the most recent aggra-dational episode within the basin (deposition of the post–0.98 Ma Quaternary Conglomer-ate) because (1) we do not have the temporal constraints on the deposition of the Alfarcito Conglomerate that are required to construct a meaningful model of the basin-fi lling event(s) during this time; and (2) the changing paleofl ow directions within the Alfarcito Conglomerate suggest that the catchment geometry was dif-ferent than that observed today, and so current basin measurements probably do not accurately represent the paleogeometry of the basin. Prior to the deposition of the Quaternary Conglomer-ate, the base level within the Toro basin (and thus at the interface between the uplifting Sierra Pasha and the basin) was close to its current base level, as the Río Toro has recently begun

to erode through the unconformity between late Miocene and Plio-Pleistocene units. The current channel geometry is thus similar to that at the commencement of the latest massive conglom-eratic deposition. If the foreland base level has remained approximately fi xed, then channel slopes within the downstream incising section and upstream aggrading section were different prior to the onset of the most recent depositional episode (Fig. 8). Therefore, we modify the model by Sobel et al. (2003) to account for the effects of preexisting topography by allowing the initial channel slopes to differ within the southern Sierra Pasha and Toro basin. Imple-menting this change, the threshold of channel defeat becomes

U

KDc=1

2SbSu

(n−1)2 πWb( )

12ka

m 1− hm( )12

Wb +Wu( )1−hm−Wb

1− hm[ ]12

, (2)

where Su and S

b are the initial channel slopes

in the southern Sierra Pasha and Toro basin, respectively. Our general approach is to use con-straints on the parameters in equation 2 to solve for the unknown values, particularly the poorly known bedrock erodibility constant, K. We then compare the value of K determined in the Toro basin within independently derived estimates in similar rock types and climatic conditions. In this way, we provide a semiquantitative test for this model to determine if its predictions are consistent with independent observations.

Constraints on Model Parameters in the Toro Basin

We constrained U, Sb, S

u, h, k

a, W

b, W

u, and

Dc in the Toro basin using geologic observations

and DEM and precipitation analyses (Table 2). First, the key elements of the model can be identifi ed in Figure 8. Upstream from the stable

Figure 9. Schematic evolution of the Toro basin during the late Miocene to the present. (A) Initial foreland basin deposition is recorded within late Miocene sediments exposed to the east of the present basin whose clast composition indicates a source from the Puna margin. (B) Changing clast composition within sediments of the Toro basin and those located to the east of the Sierra Pasha indicate that topography to the east of the Toro basin started to rise around 6.4 Ma. (C) The depositional environment, clast composition, and paleofl ow directions within the Quaternary conglomerate show rapid fi lling and probably hydrologic isolation of the Toro basin sometime after 0.98 Ma and before 40 ka. (D) Recapture and incision of the Toro basin has led to the excavation of basin sediments and erosion of paleotopography.



foreland base level is the detachment-limited portion of the channel comprised of the fault-bounded, doubly vergent wedge of metamor-phic basement rocks found in the Sierra Pasha. Presently, the downstream portions of this chan-nel are mantled by gravel deposited in a braided stream system. However, the extensive preserva-tion of strath terraces within the Río Toro gorge with thin sedimentary cover suggests that the long-term behavior of this part of the channel is incision through bedrock. Upstream of this portion of the channel are the thick Quaternary conglomerates that represent aggradation of the transport-limited section of the channel behind the rising Sierra Pasha.

Long-term rock-uplift rates within the Sierra Pasha have averaged ~0.40–0.64 mm/yr. We estimated the initial channel slopes within and behind the Sierra Pasha (S

u and S

b) to be

1.9 × 10−2 and 2.5 × 10−2, respectively. Using our high-resolution DEM combined with the Shuttle Radar Topography Mission (SRTM) DEM, we created a hydrologically corrected 90 m DEM of the entire basin using a pit-fi ll algorithm (Tarboton et al., 1991) available in the Arc/INFO GIS software that removes closed basins resulting from imprecision in the digital topographic data. We calculated channel length and upstream catchment area for each point in the DEM and performed a nonlinear regression between the two for all catchment areas greater than ~3 km2. This regression yields estimates for the power function relating area to channel length (e.g., Hack, 1957). In the case of the Toro basin, best-fi t values for k

a and h were

8.52 and 1.62, respectively (for analysis details, see Whipple, 2001; Hilley and Strecker, 2004). Using this DEM and the observed upstream location of the beginning of aggradation, we estimated the along-channel length of the trans-port- and detachment-limited sections of the channel to be W

b = 48 km and W

u = 50 km.

The fi nal parameter we constrained in equa-tion 2 was the adjusted mass diffusivity, D

c. Paola

et al. (1992) found that the absence of a critical shear-stress threshold required for sediment

Figure 10. (A) Schematic model setup in the context of the Toro basin. The channel length is divided into a downstream, detachment-limited section with length W

u, and an upstream,

transport-limited section with length Wb. S

u and S

b refer to initial channel slopes within the

uplifting basement block and headwater basin, respectively. The uplifting block is subject to a rock uplift rate of U. K and D

c are rate constants for transport in the upstream and

incision in the downstream section of the channel, respectively. Finally, basin geometric properties are encapsulated by k

a and h in the downstream portion of the channel. (B) Sche-

matic behavior of model. As the basement block undergoes uplift, a propagating knick slope will communicate the base level change between the stable Lerma Valley and the upstream areas. As surface uplift continues within the basement block, aggradation occurs within the upstream basin. Where the aggradation is insuffi cient to keep pace with the rising down-stream barrier, the channel will be defeated and internal drainage will result.

TABLE 2. MODEL PARAMETER VALUE RANGES FOR THE TORO BASIN

Model parameter Range

U 0.40–0.64 mm/yrSb 1.9 × 10–2

Su 2.5 × 10–2

h 1.62ka 8.52Wb 48 kmWu 50 kmDc 2.6 × 10–2 m/yr–

2.82 × 10–2 m/yr

TABLE 1. SYMBOLOGY USED IN TEXT

Symbol Units Description References

U m/yr Rock uplift rate within downstream detachment-limited portion of channel

Sobel et al. (2003).

K m(1–2m)/yr Bedrock erodibility constant Whipple et al. (2000), and references therein.

Dc m/yr Mass diffusivity of channels adjusted for catchment area Sobel et al. (2003); Paola et al. (1992).

So Initial channel slope prior to uplift Sobel et al. (2003).Wb m Along-channel length of upstream, transport-limited

portion of channelSobel et al. (2003).

Wu m Along-channel length of downstream, detachment-limited portion of channel traversing bedrock-cored uplift zone

Sobel et al. (2003).

ka m2-h Basin geometry scaling constant Hack (1957).h Basin geometry scaling exponent Hack (1957).m Area scaling factor in power-law incision model Howard and Kerby (1983);

Whipple and Tucker (1999); Whipple et al. (2000).

n Slope scaling factor in power-law incision model Whipple and Tucker (1999); Whipple et al. (2000).

HILLEY and STRECKER


entrainment implied a diffusion relationship between the local channel slope (S) and the volu-metric sediment fl ux in the channel (Q

s):

Qs = υS, (3)

where υ is the mass diffusivity, which is equal to:

)

υ =−8 < q < E c f

Co s−1(, (4)

where <q> is the time-averaged water discharge [m3/yr], E is a nondimensional constant that is 1 for meandering rivers and 0.2–0.4 for braided channels (Parker, 1978), c

f is the drag coeffi cient

(typically ~0.01; Paola et al., 1992), Co is the

sediment concentration, and s is the specifi c gravity of the sediment (~2.7; Paola et al., 1992). Because the only variable in equation 3 that varies with effective precipitation is <q>, we can rewrite υ in terms of <q> for the case of meandering and braided channels as (Paola et al., 1992):

υ = 0.67 <q> for meandering channels (5a)

υ = 0.10 <q> for braided channels. (5b)

To partially account for the effect of downstream increases in discharge, Sobel et al. (2003) added a catchment area term (A) to equation 3:

Qs = D

cAS, (6)

where

υ = DcA (7)

in this context. In general, discharge is com-monly related to catchment area by a power function (Dunne and Leopold, 1978):

q = cAb, (8a)

or in time-averaged variables:

<q> = <c> A<b>, (8b)

where <c> and <b> are time-averaged values of c and b. By comparison of equations 4, 6, and 8, it is apparent that one of the implicit assumptions by the formulation of Sobel et al. (2003) is that <b> = 1. This implies that no water infi ltrates in the landscape and that the time-averaged dis-charge is directly proportional to the catchment area times the time-averaged precipitation rate (<c> in this context). In the Toro basin, channel deposits indicate that braided fl uvial systems transported sediment within the basin over the

last <8 m.y. Therefore, we substitute equation 5b into 7 and 8b to fi nd that:

Dc = 0.1 <c> , (9)

where <c> is the time-averaged precipitation rate [m/yr]. Therefore, we can place bounds on the adjusted mass diffusivity knowing the time-averaged precipitation rate, <c>.

We can place some constraints on <c> based on the current distribution of precipitation in the area. First, the magnitude of the mean annual precipitation in the area is strongly controlled by the distribution of topography (Fig. 2). The humid foreland contains relatively few ranges that intercept predominantly east-northeastward derived precipitation, resulting in ~1.2 m/yr of mean annual precipitation. Conversely, the topography of the Sierra Pasha reduces precipi-tation on its leeward side, leading to an average of 0.260 m/yr of precipitation within the Toro basin (WMO, 1975; Bianchi and Yañez, 1992; Haselton et al., 2002). Farther to the south, uplift of the windward ranges has led to the progres-sive aridifi cation of leeward basins (Kleinert and Strecker, 2001). Likewise, the transition from a more humid climate recorded by the fern-bear-ing fossil units at the base of the Alfarcito Conglomerate to the arid conditions indicate an overall drying trend in the Toro basin over the last several m.y. Therefore, a minimum value for D

c likely lies around 0.0260 m/yr.

An upper bound on Dc can be estimated

from the thickness of the sediments within the Toro basin that accumulated behind the rising downstream topography. Field observations and DEM measurements indicate that 600 m of incision has occurred between the highest preserved geomorphic surface and the current channel elevation within the basin (section 3). In the model by Sobel et al. (2003), the time at which hydrologic isolation occurs is:

t =1

4Sb2U −2πDcWb. (10)

The thickness of the sediment accumulated directly behind the rising barrier at this time, T

s,

is equal to the uplift rate times this time. Substi-tuting t = T

s/U into equation 9, we fi nd that D

c

may be expressed in terms of Ts, S

b, U, and W

b:

Dc = 4T

sS

b–2UπW

b–1. (11)

The geomorphic character and degree of varnishing of clasts on the smooth preserved remnants of the highest geomorphic surface in the Toro basin suggest that they have undergone only minimal erosional lowering since abandon-ment and incision (Strecker et al., 2002). There-fore, the total thickness of the fi ll units (T

s) at

the downstream end of the aggrading zone was ~600 m above the current channel elevation. Using this value for T

s, U = 0.40–0.64 mm/yr,

Wb = 48 km, and S

b = 0.019, we estimate a maxi-

mum value for Dc to be 0.0282 m/yr. Using the

minimum and maximum bounds for Dc and S

b,

we used equation 9 to calculate the time that it should take the basin to become hydrologically isolated once aggradation began. The model requires between 620–672 k.y. to aggrade to a point where internal drainage is established within this basin. This amount of time is con-sistent with the 0.98 Ma maximum age of the Quaternary Conglomerate that represents the last basin-fi lling event we model.

Model Results

Using estimated values for ka, h, W

b, and

Wu, we calculated the threshold of hydrologic

isolation for various combinations of m and n (Sobel et al., 2003; Whipple et al., 2000). For the observed geometry of the Toro basin, we used equation 2 to estimate U KDc values of 0.89, 0.94, and 310 when m = 1/3 and n = 2/3, m = 0.4 and n = 1, and m = 5/4 and n = 5/2, respec-tively, to maintain a basin close to its threshold of hydrologic isolation. Next, we used the range for U and D

c values to estimate the unknown

rock erodibility factor, K. Inferred K values are 1.2 ± 0.8 × 10−5, 1.1 ± 0.7 × 10−5, and 9.7 ± 6.5 × 10−11 when m = 1/3 and n = 2/3, m = 0.4 and n = 1, and m = 5/4 and n = 5/2, respectively.

DISCUSSION

Time Scales of Processes Leading to Basin Filling and Excavation

The stratigraphy of the Toro basin shows that different processes may lead to the fi lling and reexcavation of basin material over various time scales. The Alfarcito and post–0.98 Ma Quater-nary Conglomerate record at least two cycles of fi lling and reexcavation. During the fi rst episode, fi lling of the basin proceeded over <7 m.y., pro-ducing a sedimentary fi ll whose thickness was certainly greater than 600 m. In a relatively short time (<0.98 m.y.), the trunk stream eroded to a level similar to the present base level and mate-rial was removed from the basin. Subsequently, this erosional paleotopography was infi lled to levels ~600 m above the trunk stream. In turn, this fi ll was episodically removed and the drain-age became fully integrated with the foreland. Therefore, deposition and erosion cycles in this tectonic environment that involve large volumes of sediments (>61.5 km3) may occur over time scales of 1 m.y. or less, and may oscillate on similar time scales.



In contrast, shorter time-scale events of fi ll-ing and excavation are also recorded within the Quaternary strata of the Toro basin. Trauth and Strecker (1999) describe an onlapping lacustrine deposit that formed behind a landslide dam ca. 30 ka. This landslide may have been triggered by brief periods of intensifi ed precipitation, runoff, and lateral scouring (Trauth and Strecker, 1999); however, other large landslides in NW Argentina may also have been triggered by seismic activity (Hermanns and Strecker, 1999). The landslide dam in the Río Toro gorge was removed shortly after it formed, resulting in the evacuation of most of the lacustrine sediments and associated strata. Therefore, while the short time-scale damming events driven by extreme climatic events or seismicity may lead to high transient aggradation rates, their long-term impact on the basin history may be overwhelmed by the effects of downstream uplift, aggradation, and recapture of the trunk stream at the outlet.

Model Results

The inferred values of the incision rate con-stant (K) allow us to provide a semiquantitative test for the model of Sobel et al. (2003) by comparing these values to those determined independently in similar rock types. First, the most comprehensive analysis of the incision of fl uvial systems into bedrock found that the fi rst-order control on K was rock type (Stock and Montgomery, 1999). This study found that for areas in which the bedrock incision power-law exponents (m and n) could be constrained, values of m = 0.4 and n = 1 yielded the best representation of the rates and geometries of the channel profi les over time (Stock and Mont-gomery, 1999). For these exponents, K values for granitoid/metasedimentary rocks and vol-caniclastic rocks were inferred to be 4.4 × 10−7 to 4.3 × 10−6, and 4.8 × 10−5 to 3.0 × 10−4, for sites in Australia and California, respectively. In addition, estimates of K within the Siwalik molasse, Nepal, are between 1.47 × 10−4 and 1.64 × 10−4 (Kirby and Whipple, 2001). Finally, Hilley and Strecker (2004) used a coupled tectonic-erosional model to estimate K values between 1.5 × 10−5 to 2.5 × 10−5 and 2.9 × 10−7 to 2.3 × 10−7 for the sedimentary and metamor-phic rocks exposed in Taiwan and the Himalaya, respectively. In this study, we estimate K to be 1.1 ± 0.7 × 10−5 in the southern Sierra Pasha. While estimated K values for a given rock type appear to vary over approximately an order of magnitude, differences between rock types may vary up to four orders of magnitude (Stock and Montgomery, 1999; Kirby and Whipple, 2001). Hence, the low-grade, fractured metamorphic rocks found within this area yield K values that

are consistent with those determined elsewhere for similar rock types. Based on these compari-sons, it appears that this simple model may cap-ture the essence of channel defeat and the ensu-ing hydrologic isolation of the basins behind the rising barrier. Therefore, in intramontane basins that undergo aggradation by similar processes, this model may provide insight as to the hydrologic, tectonic, climatic, and lithologic conditions necessary to isolate them from the foreland fl uvial system.

There are several important simplifi cations of the processes leading to the formation of internal drainage in the formulation of Sobel et al. (2003) and our modifi ed model that warrant discussion. First, the model idealizes aggrada-tion upstream of the rising topography as a diffusion process whose rate constant (υ) scales continuously with the time-averaged discharge and hence the basin area. However, numerous tributaries along the Río Toro cause discharge to increase rapidly where they meet the trunk stream. While these discharge point sources are not explicitly treated in this formulation, their abundance leads to an approximation of a continuous downstream increase in discharge. Second, our calculations do not model a precipi-tation gradient within the basin but instead use a constant, basin-wide average. In the case of our study area, the basin parallels a high range to the east, and precipitation gradients are small across the area. However, in wide basins or those with signifi cant along-length topography, down-stream changes in time-averaged precipitation may complicate the channel defeat threshold encapsulated in equation 1. Third, an implicit assumption of the nonlinear diffusion law in alluvial rivers is that the shear stress required for sediment entrainment is small relative to the shear stress exerted on the channel bed (Paola et al., 1992). Because no entrainment shear stress is included, the potentially impor-tant role of clast-size variations through time (perhaps driven by changing tectonic activity or climatic conditions) is neglected. In the early stages of aggradation, relatively steep channel slopes may cause the channel bed shear stress to greatly exceed the entrainment stress. However, as aggradation proceeds and channel slopes shallow, the channel-bed shear stress may begin to approach the entrainment shear stress, which may reduce transport capacity, favor aggrada-tion, and cause the channel to be defeated earlier than the model predicts. It is important to note that within the Toro basin, there is an increase in average clast size within the post–0.98 Ma Quaternary Conglomerate compared to the underlying Alfarcito Conglomerate, and so this effect may have partially aided aggradation in the basin. While the incorporation of clast-size

variations into an internal drainage threshold such as equation 1 is an important problem and should be the subject of further research, the consistency of the incision-rate constants inferred within the Toro with those determined elsewhere suggests that this simple model captures many of the fi rst-order features of aggrading basins. Finally, because the bedrock power-law incision model, used in the detach-ment-limited portion of the modeled channel, is not based on a specifi c physical process (such as Sklar and Dietrich, 1998), but rather an energy-expenditure hypothesis (Seidl and Dietrich, 1992), the comparison of results with controlled experiments may be diffi cult (Sklar and Dietrich, 2001). As the determination of specifi c bedrock incision mechanisms become better formalized, a more physically based inci-sion rule may come to light. However, Whipple and Meade (2004) argued that the complexities of individual bed-erosion mechanisms and their interaction may be captured by an empirical power-law scaling rule. Therefore, while simple and empirical, the bedrock power-law incision model may provide enough fl exibility to capture the development of incising fl uvial systems over large spatial and temporal scales.

Tectonic and Climatic Infl uences on Basin Filling

Observations from and models of the Toro basin imply that a complex relationship between rock uplift, rock type, and climate may control the oscillatory character of basin development that is also observed in other parts of the Andes or other orogens (e.g., Beer et al., 1990; Upton et al., 2003). However, the pervasiveness and simi-larity of these basins suggests that a common set of processes is responsible for their existence. In the case of the Toro basin and other intra-montane basins along the eastern Puna margin (Strecker et al., 1989; Strecker et al., 2002), the establishment of orographic barriers by surface uplift concentrates erosion along the windward eastern range fronts, whereas the leeward basins become increasingly arid (Kleinert and Strecker, 2001). Basin outlets in this environment always coincide with structurally complex parts of the orogen where along-strike discontinuities allow the fl uvial system to remain connected to the foreland. In addition, these sectors of the orogen coincide with high precipitation gradients that may aid in maintaining external drainage condi-tions where moisture is funneled into the inte-rior of the orogen. However, changes in climate and hence variability in runoff and sediment, tectonic rates, or the unroofi ng of resistant units may lead to reduced evacuation of sediment and potentially to basin isolation.

HILLEY and STRECKER


Our model provides a means of estimating the potential contribution of rock uplift, precipi-tation, and exposed rock type on the hydrologic isolation of intramontane basins by considering a range of these values appropriate for actively deforming orogens. First, rock uplift rates may range over three orders of magnitude between 0.1–10 mm/yr (e.g., Sobel and Strecker, 2003). Precipitation typically varies over two orders of magnitude between 100–10,000 mm/yr (WMO, 1975; Bianchi and Yañez, 1992), whereas rock resistance to fl uvial incision varies over four orders of magnitude (Stock and Montgomery, 1999). In our model formulation, the threshold of hydrologic isolation is defi ned by U KDc, with D

c varying directly with time-averaged pre-

cipitation. Given the ranges in possible values for U, K, and D

c, our model suggests that the

relative importance of these factors (in decreas-ing order) is rock uplift, resistance of bedrock to fl uvial incision, and precipitation. However, these parameters may not be independent, as the construction of topography modifi es the precipi-tation (e.g., Roe et al., 2002, 2003) and perhaps the distribution of deformation and hence range-scale rock uplift rates (e.g., Davis et al., 1983; Dahlen and Suppe, 1988; Willett et al., 1993; Willett, 1999; Schlunegger and Simpson, 2002; Upton et al., 2003) within the orogen.

Careful geomorphic, structural, and sedi-mentologic observations may help to test the theoretical importance of rock uplift, climate, and exposed rock types in the evolution of these basins. In the case that climatic fl uctuations strongly control these processes, we expect the timing of intramontane basin fi lling and excava-tion to be synchronous over several degrees of latitude. However, rock uplift rates may vary over much smaller length scales, and if this is the dominant control, we may expect basin fi ll-ing and excavation to be diachronous between individual basins. A series of at least fi ve large intramontane basins situated along the margin of the Puna contain abundant evidence of cyclic fi lling and reexcavation over the last several million years (Strecker et al., 1989; Strecker et al., 2002; Carrapa et al., 2004). Therefore, a high-resolution documentation of these recur-ring processes may eventually provide a means of testing the relative importance of tectonic versus climatic forcing factors in controlling basin history.

CONCLUSIONS

1. Sediments within the Toro basin record a transition between a continuous foreland depo-sitional environment and a compartmentalized foreland sometime between 8 and 6 Ma. The foreland was segmented by the uplift of the

Sierra Pasha to the east, which reorganized regional drainage and initiated erosion fol-lowed by aggradation within the basin that lasted until at least 4.17 Ma. Between this time and 0.98 Ma, deformation shifted into the Toro basin, resulting in the deformation of underly-ing units and deposition of syntectonic sedi-ments. After 0.98 Ma, the basin was once again excavated, and a currently undeformed coarse conglomeratic unit was deposited. Finally, these basin sediments were episodically reexcavated. Thus, this area has experienced at least two cycles of fi lling and reexcavation in Plio-Pleis-tocene time, and these processes appear to be a long-lived feature of this basin.

2. Geologic observations suggest that chan-nel steepening driven by uplift within the Sierra Pasha Sur has caused the most recent aggrada-tion event and probably previous aggradation episodes. The sedimentary provenance and geometry of Quaternary erosion surfaces sug-gest that uplift within the Sierra Pasha and the Sierra Pasha Sur either pushed the Río Toro close to or over its threshold of defeat. This resulted in basin fi lling and possibly the hydro-logic isolation of this basin in the last 0.98 m.y.

3. Using our geologic and geomorphic observations, a synthesis of published geologic data, and DEM analysis, we constrained the parameters of a modifi ed version of a model of the establishment of internal drainage by downstream channel steepening proposed by Sobel et al. (2003). Using this model, we solved for the poorly constrained bedrock incision rate constant, K, within the Sierra Pasha Sur required to bring the Río Toro close to its threshold of defeat. We infer K values for the fractured and weakly metamorphosed metasediments of the Sierra Pasha Sur to be 3.8 ± 1.2 × 10−5, 3.2 ± 1.0 × 10−5, and 1.6 ± 0.51 × 10−10 when m = 1/3 and n = 2/3, m = 0.4 and n = 1, and m = 5/4 and n = 5/2, respec-tively. These estimates fall between those for metasediments/granitoids and volcaniclastics/mudstones presented by Stock and Montgom-ery (1999) for m = 0.4, n = 1. While there is no direct lithologic equivalent for which these val-ues have been estimated, our range determined for rocks of the Sierra Pasha Sur are qualita-tively reasonable. Therefore, given the inherent uncertainties in the erosional exponents and incision constant in the stream-power bedrock incision model, the simple model of Sobel et al. (2003) seems to capture the general pro-cesses of fi lling and perhaps the establishment of internal drainage observed within the Toro basin and may be used to understand these processes within other basins with similar his-tories along the Puna margin in NW Argentina and other orogens.

ACKNOWLEDGMENTS

We would like to thank R. Alonso for his scientifi c expertise and participation in several fi eld excursions to the study area. GH acknowledges the Alexander von Humboldt Foundation and the International Quality Network (IQN) for funding his postdoctoral research in Germany. MS acknowledges support from the Spe-cial Research Project 267 (Deformation Processes in the Andes) of the German Research Council. Conver-sations with I. Coutand, E. Sobel, and M. Trauth have contributed to the ideas presented herein. We thank R. Allmendinger, T. Demko, and M. Gerdes for valuable comments on the fi rst draft of this manuscript.

REFERENCES CITED

Alonso, R.N., and Marquillas, R.A., 1978, El Grupo Mesón: Contenido paleontologico y edad: Acta Geologica Lil-loana, v. 14, p. 5–6.

Beer, J.A., Allmendinger, R.W., Figueroa, D.E., and Jordan, T.E., 1990, Seismic stratigraphy of Neogene piggyback basin, Argentina: American Association of Petroleum Geologists Bulletin, v. 74, p. 1183–1202.

Bianchi, A.R., and Yañez, C.E., 1992, Las precipitaciones en el noroeste Argentino: Salta, Instituto Nacíonal de Tec-nología Agropecuaria, Estacíon Agropecuaria, 35 p.

Carrapa, B., Strecker, M.R., and Sobel, E., 2004, Sedimen-tary, tectonic and thermochronologic evolution of the southernmost end of the Puna Plateau (NW Argen-tina): The intramontane Bolsón de Fiambalá, Interna-tional geological congress (IGC): Florence Electronic abstract 147-26, session G.23.01.

Dahlen, F.A., and Suppe, J., 1988, Mechanics, growth, and erosion of mountain belts: Geological Society of America Special Paper 218, p. 161–178.

Davis, D., Suppe, J., and Dahlen, F.A., 1983, Mechanics of fold-and-thrust belts and accretionary wedges: Journal of Geophysical Research, v. 88, p. 1153–1172.

DeCelles, P.G., and Giles, K.A., 1996, Foreland basin sys-tems: Basin Research, v. 8, p. 105–123.

DeCelles, P.G., Gray, M.B., Ridgway, K.D., Cole, R.B., Srivastava, P., Pequera, N., and Pivnik, D.A., 1991, Kinematic history of foreland uplift from Paleocene synorogenic conglomerate: Beartooth Range, Wyo-ming and Montana: Geological Society of America Bulletin, v. 103, p. 1458–1475.

Dunne, T., and Leopold, L.B., 1978, Water in environmental planning: New York, W.H. Freeman and Company, 818 p.

Hack, J.T., 1957, Studies of longitudinal stream profi les in Virginia and Maryland: U.S. Geological Survey Pro-fessional Paper 294, 97 p.

Haselton, K., Hilley, G.E., and Strecker, M.R., 2002, Average Pleistocene climatic patterns in the southern central Andes: Controls on mountain glaciation and paleoclimate implications: Journal of Geology, v. 110, p. 211–226.

Hermanns, R.L., and Strecker, M.R., 1999, Structural and lithologic controls on large Quaternary rock avalanches (sturzstroms) in arid northwestern Argen-tina: Geological Society of America Bulletin, v. 111, p. 934–948.

Hilley, G.E., and Strecker, M.R., 2004, Steady-state ero-sion of critical Coulomb wedges with applications to Taiwan and the Himalaya: Journal of Geophysical Research, v. 109, B01411.

Howard, A., and Kerby, G., 1983, Channel changes in bad-lands: Geological Society of America Bulletin, v. 94, p. 739–752.

Humphrey, N.F., and Konrad, S.K., 2000, River incision or diversion in response to bedrock uplift: Geology, v. 28, p. 43–46.

Hutchinson, M.F., 1989, A new procedure for gridding elevation and stream line data with automatic removal of spurious pits: Journal of Hydrology, v. 106, p. 211–232.

Isacks, B., 1988, Uplift of the Central Andean Plateau and bending of the Bolivian orocline: Journal of Geophysi-cal Research, v. 93, p. 3211–3231.



Jordan, T.E., Allmendinger, R.W., Damanti, J.F., and Drake, R.E., 1993, Chronology of motion in a complete thrust belt: The Precordillera, 30–31°S, Andes Mountains: Journal of Geology, v. 101, p. 137–158.

Kirby, E., and Whipple, K.X., 2001, Quantifying differential rock uplift rates via stream profi le analysis: Geology, v. 29, p. 415–418.

Kleinert, K., and Strecker, M.R., 2001, Changes in moisture regime and ecology in response to late Cenozoic oro-graphic barriers: The Santa Maria Valley, Argentina: Geological Society of America Bulletin, v. 113, p. 728–742.

Marrett, R.A., and Allmendinger, R.W., 1990, Kinematic analysis of fault slip data: Journal of Structural Geology, v. 12, p. 973–986.

Marrett, R.A., and Strecker, M.R., 2000, Response of intracontinental deformation in the central Andes to late Cenozoic reorganization of South American plate motions: Tectonics, v. 19, p. 452–467.

Marrett, R.A., Allmendinger, R.W., Alonso, R.N., and Drake, R.E., 1994, Late Cenozoic tectonic evolution of the Puna Plateau and adjacent foreland, northwestern Argentine Andes: Journal of South American Earth Sciences, v. 7, p. 179–207.

Mikuz, T., 2002, Die Geologie der Sierra de Pasha in der Ostkrdillere von Nordwest Argentinien [M.S. thesis]: Vienna, Austria, Universität Wien, 124 p.

Paola, C., Heller, P.L., and Angevine, C.L., 1992, The large-scale dynamics of grain-size variation in alluvial basins: I: Theory: Basin Research, v. 4, p. 73–90.

Parker, G., 1978, Self-formed straight rivers with equilib-rium banks and mobile bed, part II, the gravel river: Journal of Fluid Mechanics, v. 89, p. 127–148.

Reyes, F.C., and Salfi ty, J.A., 1973, Consideraciones sobre la estratigrafía del Cretácico (Subgrupo Pirgua) del noroeste Argentino: Actas Quinto Congrreso Geológico Argentino v. 3, p. 355–385.

Reynolds, J.H., Galli, C.I., Hernández, R.M., Idleman, B.D., Kotila, J.M., Hilliard, R.V., and Naeser, C.W., 2000, Middle Miocene tectonic development of the transition zone, Salta Province, northwest Argentina: Magnetic stratigraphy from the Metán Subgroup: Sierra de González: Geological Society of America Bulletin, v. 112, p. 1736–1751.

Roe, G.H., Montgomery, D.R., and Hallet, B., 2002, Effects of orographic precipitation variations on the concav-ity of steady-state river profi les: Geology, v. 30, p. 143–146.

Roe, G.H., Montgomery, D.R., and Hallet, B., 2003, Oro-graphic precipitation and the relief of mountain ranges: Journal of Geophysical Research, v. 108.

Schlunegger, F., and Simpson, G., 2002, Possible erosional control on lateral growth of the European Central Alps: Geology, v. 30, p. 907–910.

Schwab, K., and Schäfer, A., 1976, Sedimentation und Tektonik im mittleren Abschnitt des Río Toro in der Ostkordillere NW-Argentiniens: Geologische Rund-schau, v. 65, p. 175–194.

Seidl, M., and Dietrich, W.E., 1992, The problem of chan-nel erosion into bedrock: Functional Geomorphology, Catena, supplement 23, p. 101–124.

Sklar, L.S., and Dietrich, W.E., 1998, River longitudinal pro-fi les and bedrock incision models; stream power and the infl uence of sediment supply, in Tinkler, K.J., and Wohl, E.E., eds., Rivers Over Rock: Fluvial Processes in Bedrock Channels: Washington, D.C., American Geophysical Union, p. 237–260.

Sklar, L.S., and Dietrich, W.E., 2001, Sediment and rock strength controls on river incision into bedrock: Geol-ogy, v. 29, p. 1087–1090.

Sobel, E.R., and Strecker, M.R., 2003, Uplift, exhuma-tion, and precipitation: Tectonic and climatic control of late Cenozoic landscape evolution in the northern Sierras Pampeanas, Argentina: Basin Research, v. 15, p. 431–451.

Sobel, E.R., Hilley, G.E., and Strecker, M.R., 2003, Forma-tion of internally drained contractional basins by arid-ity-limited bedrock incision: Journal of Geophysical Research, v. 108.

Starck, D., and Anzótegui, L.M., 2001, The late Miocene cli-mate change—persistence of a climatic signal through the orogenic stratigraphic record in northwestern Argentina: Journal of South American Earth Sciences, v. 14, p. 763–774.

Stock, J.D., and Montgomery, D.R., 1999, Geologic con-straints on bedrock river incision using the stream power law: Journal of Geophysical Research, Solid Earth, v. 104, p. 4983–4993.

Strecker, M.R., Cerveny, P., Bloom, A.L., and Bagemann, F., 1989, Late Cenozoic tectonism and landscape develop-ment in the foreland of the Andes; northern Sierras Pam-peanas (26°–28°), Argentina: Tectonics, v. 8, p. 517–534.

Strecker, M.R., Hilley, G.E., and Trauth, M.H., 2002, Tec-tonic and climatic control on transitory intramontane basin fi lling in the southern central Andes: The Toro basin at the eastern Puna margin, NW Argentina: Geophysical Research Abstracts [CD-ROM], Abstract EGS02-A-06063.

Suppe, J., Chou, G.T., and Hook, S.C., 1992, Rates of fold-ing and faulting determined from growth strata, in McClay, K.R., ed., Thrust tectonics: London, Chapman and Hall, p. 105–122.

Tarboton, D.G., Bras, R.L., and Rodriguez-Iturbe, I., 1991, On the extraction of channel networks from digital ele-vation data: Hydrological Processess, v. 5, p. 81–100.

Trauth, M.H., and Strecker, M.R., 1999, Formation of landslide-dammed lakes during a wet period between 40,000–25,000 yr B.P. in northwestern Argentina: Paleogeography, Palaeoclimatology, Palaeoecology, v. 153, p. 277–287.

Upton, P., Mueller, K.J., Koons, P.O., and Powell, L., 2003, Reorganization of strain in response to erosional forc-ing at intermediate scales: Puli Embayment, Western Taiwan: Eos (Transactions, American Geophysical Union), v. 84(46) Fall Meeting supplement, Abstract.

Viramonte, J.G., Reynolds, J.H., Del Papa, C., and Disalvo, A., 1994, The Corte Blano garnetiferous tuff: A distinctive late Miocene marker bed in northwestern Argentina applied to magnetic polarity stratigraphy in the Río Yacones: Salta Province: Earth and Planetary Science Letters, v. 121, p. 519–531.

Whipple, K.X., 2001, Fluvial landscape response time: How plausible is steady-state denudation?: American Jour-nal of Science, v. 301, p. 313–325.

Whipple, K.X, and Meade, B.J., 2004, Controls on the strength of coupling among climate, erosion, and deformation in two-sided, frictional orogenic wedges at steady-state: Journal of Geophysical Research, v. 109, F01011.

Whipple, K.X., and Tucker, G.E., 1999, Dynamics of the stream-power river incision model: Implications for height limits of mountain ranges, landscape response time scales, and research needs: Journal of Geophysi-cal Research, v. 104, p. 17,661–17,674.

Whipple, K.X., Hancock, G.S., and Anderson, R.S., 2000, River incision into bedrock: Mechanics and relative effi cacy of plucking, abrasion, and cavitation: Geologi-cal Society of America Bulletin, v. 112, p. 490–503.

Willett, S.D., 1999, Orogeny and orography: The effects of erosion on the structure of mountain belts: Journal of Geophysical Research, v. 104, p. 28,957–28,981.

Willett, S.D., Beaumont, C., and Fullsack, P., 1993, Mechan-ical model for the tectonics of doubly vergent compres-sional orogens: Geology, v. 21, p. 371–374.

World Meteorological Organization (WMO), 1975, Climatic atlas of South America: World Meteorological Organi-zation, Geneva, scale 1:1,000,000, 28 p.

MANUSCRIPT RECEIVED BY THE SOCIETY 27 FEBRUARY 2004REVISED MANUSCRIPT RECEIVED 13 OCTOBER 2004MANUSCRIPT ACCEPTED 11 NOVEMBER 2004

Printed in the USA

Documents

Processes of oscillatory basin ﬁ lling and excavation in a ...pangea.stanford.edu/.../HilleyandStrecker_GSAB2005.pdf · rocks. In the vicinity of the Toro basin, uplifted basement