The ‘Rodados Patagónicos’ (Patagonian shingle formation) of eastern Patagonia: environmental conditions of gravel sedimentation

Authors


E-mail: oscarm@unpata.edu.ar

Abstract

The gravel mantels known as ‘Rodados Patagónicos’ of Eastern Patagonia are one of the most distinctive features of the regional landscape. Their significant roundness, along with additional sedimentary and geomorphological evidence, indicates their origin is linked to fluvial action. Currently, no natural processes exist on Earth potentially generating equivalent deposits, either in Patagonia or elsewhere around the world. The hydrological conditions responsible for the gravel formation were thus likely very different from those currently prevailing in Argentine Patagonia, suggesting the gravel mantels formed during the Late Cenozoic, when surface drainage networks of relatively high energy existed in the region. Such high energy was induced by powerful flow rates and/or strong water gradients, attributable to tectonism, epirogenesis and/or lowering of sea level. Periods with water flow regimes significantly higher than at present characterized the full glacial phases lasting several thousands of years, corresponding to each of the many glaciations of the Pliocene and Pleistocene. The capacity for erosion and water transport would have been increased by the sea level drop characteristic of each full glacial phase. There were also much shorter periods, known as ‘terminations’, representing the abrupt ending of the main glacial events during which large amounts of water were released as a result of the intensive melting of the Cordilleran ice sheet. During interglacial periods, in the absence of large masses of ice in the mountains and with average temperatures similar to or higher than at present, layers of gravel were deposited as a result of tectonic or epirogenic movement or by reactivation of drainage networks produced by sea level lowering (glacio-eustatism). These endogenous landscape reactivations were also independent of and longer than the climatic fluctuations and therefore overlapped with both interglacial and glacial periods. Finally, some units of the ‘Rodados Patagónicos’ may also have been deposited during major pluvial events, characterized by higher mean annual precipitation over sufficiently long periods. The current evidence suggests that, at various times since the Late Miocene, varied climatic conditions have allowed the production and accumulation of large gravel beds covering much of extra-Andean Patagonia. © 2011 The Linnean Society of London, Biological Journal of the Linnean Society, 2011, 103, 336–345.

Los mantos de gravas de Patagonia oriental, conocidos como ‘Rodados Patagónicos’, constituyen por su extensión y homogeneidad uno de los elementos del paisaje más distintivos de la Patagonia oriental. El marcado redondeamiento de estos sedimentos, junto con otras evidencias sedimentológicas y geomorfológicas, indica que su origen se vincula inexorablemente con el escurrimiento superficial (acción fluvial). Puesto que en la actualidad no existen procesos que estén generando depósitos o geoformas equivalentes (en Patagonia o en otro sitio de la superficie terrestre), debe asumirse que las condiciones hídricas que permitieron su formación fueron muy distintas a las que imperan actualmente en la Patagonia Argentina. Es necesario considerar entonces ciertos momentos dentro del Cenozoico Superior, en los que existieron redes de drenaje superficial con mayor energía, inducida ésta por mayores caudales y/o mayores gradientes hídricos (por tectónica, epirogénesis o descenso del nivel del mar). Lapsos con regímenes hídricos y caudales significativamente mayores a los actuales caracterizaron las fases pleniglaciales (períodos de estabilización y máxima expansión de los glaciares que duraron quizás decenas de miles de años) correspondientes a cada una de las numerosas glaciaciones ocurridas durante el Plioceno y el Pleistoceno. En estos casos la capacidad de erosión y transporte de las aguas superficiales se habría visto incrementada por el descenso del nivel del mar, lo cual tiene lugar en cada episodio glacial. Durante estos eventos climáticos las condiciones fueron típicamente glaciales (muy frías y húmedas) en el ámbito cordillerano y periglaciales (muy frías y secas) en Patagonia extraandina. Por otro lado, existieron momentos mucho más breves, denominados ‘terminaciones’, que coincidieron con la brusca finalización de las glaciaciones, en los cuales se liberaron grandes cantidades de agua provenientes del intenso derretimiento del manto de hielo cordillerano. Este espectacular proceso se habría producido bajo condiciones climáticas similares al actual interglacial. Por otro lado, durante los períodos interglaciales, en ausencia de las grandes masas de hielo en la cordillera y con temperaturas medias atmosféricas similares o superiores a las actuales, se depositaron mantos de grava pedemontanos como producto de ascensos/descensos tectónicos o epirogénicos (cordilleranos y extracordilleranos) o por la reactivación de las redes de drenaje producida por el descenso glacioeustático. Estas reactivaciones del paisaje, de carácter endógeno, fueron independientes de las fluctuaciones climáticas y de mayor duración que éstas y, por lo tanto, actuaron durante el tiempo superponiéndose a los interglaciales y a las glaciaciones, indistintamente. Finalmente, no debería descartarse la posibilidad que algunas unidades de rodados hayan sido depositadas durante eventos pluviales mayores, los que implicaron un aumento de las precipitaciones medias anuales durante un lapso suficientemente prolongado.

La evidencia con que se cuenta en la actualidad indica que, en distintos momentos desde el Mioceno Tardío, condiciones climáticas muy variadas habrían favorecido o, al menos, permitido la producción y acumulación de grandes mantos de gravas, muy similares entre sí, que cubrieron buena parte de la superficie patagónica extraandina. Las más extensas de estas unidades sedimentarias se habrían depositado en contextos ecológicos también muy variados, extendiéndose por toda Patagonia atravesando ambientes de bosque, tundra, estepas graminosas y arbustivas, y zonas de transición entre las formaciones vegetales mencionadas.

INTRODUCTION

Patagonia is likely to have started experiencing a process of desertification at approximately 16.5 Myr as a result of the Andean uplift (Stern & Blisniuk, 2002), with the severity of this process increasing at approximately 14 Myr (Blisniuk et al., 2005), when a new tectonic pulse increased the efficiency of the orographic effect, limiting the incoming humid winds from the Southern Pacific Ocean (Ramos & Ghiglione, 2008). This process generated a marked west–east climatic gradient similar to the one that characterizes present-day Patagonia east of the Andes (e.g. mean annual precipitation 5000 mm in the Andean environment and < 200 mm less than 200 km to the east at 43°S). Extensive gravel sheets known as ‘Rodados Patagónicos’ or ‘Patagonian Shingle Formation’ were deposited in eastern Patagonia during this period and under this general climatic pattern superimposed on the glacial cycles of the Quaternary. The ‘Rodados Patagónicos’ are gravel accumulations, with or without carbonate cement, substantially rounded, with pebbles and cobbles as the dominant size fractions, in a sandy or silty/clayish matrix (Martinez, Rabassa & Coronato, 2009). The larger clasts are of highly variable lithology, although with a certain predominance of basic and mesosilicic volcanics and acid plutonic rocks. Their range in Argentine Patagonia extends from the Andean Cordillera to the Atlantic Ocean and from the Río Colorado valley to the island of Tierra del Fuego (Fig. 1). They were generated during the Late Cenozoic, tend to form horizontal to subhorizontal mantles of varied extension and thickness and are located in different topographical positions, usually showing a west–east dominant gradient. Many of these layers of gravel are connected to the west with Pleistocene marginal moraines, a fact that assigns them an unquestionable glaciofluvial origin and facilitating in many cases precise ageing. However, there are also gravel accumulations of clear piedmont origin across Patagonia, an origin indicated by their spatial connection with mountain fronts, their geomorphological context and lithological affinities with the source area. The ages of these deposits are highly variable. Finally, there are ‘Rodados Patagónicos’ units, similar to those above, but which have been disconnected from old glacial margins or mountain fronts, making their genetic and temporal interpretation difficult.

Figure 1.

The range of ‘Rodados Patagónicos’ in Argentine Patagonia between the Andean Cordillera and the Atlantic Ocean coast (Martínez & Coronato, 2008).

This paper provides an analysis of various environmental conditions under which these unique depositional sedimentary units were likely formed. Our analysis assumes that different geomorphological processes during the Late Cenozoic were capable of generating very similar landscape deposits and forms at different times and under different ecological contexts (Martínez & Coronato, 2008).

PALAEOENVIRONMENTAL CONDITIONS AND MODERN VEGETATION IN EXTRA-ANDEAN PATAGONIA

The physiognomy of the vegetation is a precise climate indicator, as shown by the different lifestyles that characterize the world's biomes (Hinojosa, 2004). Volkheimer (1971) analysed the relationship between leaf morphological characters and climatic variables (mainly temperature and rainfall) during the Cenozoic of southern South America and proposed a progressive decrease in temperature, from subtropical to temperate, and the establishment of arid to semi-arid conditions in the area towards the Pliocene (Hinojosa, 2004). A series of factors have affected this increase in aridity in the subtropics of South America (Hinojosa & Villagrán, 1997; Villagrán & Hinojosa, 1997). These include the separation of South America from Antarctica, the subsequent glaciation of West Antarctica and the generation of the Circumpolar and Humboldt currents in their present forms (Simpson, 1983; Hinojosa & Villagrán, 1997; Villagrán & Hinojosa, 1997). The interaction of these factors, coupled with the rain shadow effect generated by the uprising of the Andes, ending around the Pliocene/Pleistocene boundary, are likely to have forced the fragmentation of Subtropical Palaeoflora and the expansion of the arid taxa along what is called the ‘Arid Diagonal’. This arid diagonal is a continuous strip of arid climate that extends along the Andes from western Venezuela and north-western Chile, Argentina to north-eastern Patagonia (Villagrán & Hinojosa, 1997, 2005; Ezcurra, 2002; Hinojosa, 2004), which in turn coincides with the South American transition zone. In this zone, biotic elements of the Andean and Neotropical regions overlap (Morrone, 2001, 2004; Ruggiero & Ezcurra, 2003) and its formation can be considered a vicarious event that has slowed the spread of Andean and Neotropical biotas (Morrone, 2004). Additionally, the xerophytic vegetation along this stretch isolated the forest region of southern South America from the other forest formations of the continent (Villagrán & Hinojosa, 2005).

Currently, southern South America is characterized by a marked climatic and vegetation contrast between both sides of the Andes. Particularly, the vegetation of Andean Patagonia occupies most of the Patagonian territory, occurring as a semi-desert, with alternating grass steppes in the western sectors of higher moisture, with shrub and subshrub steppes to the east (Soriano, 1956; Cabrera & Willink, 1980). From a phytogeographical point of view, Patagonia includes elements of the Neotropical region, the Andean Patagonian Domain and the Patagonian province. The Patagonian province is itself divided into the Western, Central and Sub-Andean Districts (Soriano, 1956).

These environments are characterized by a homogeneous appearance, with extensive plains some with slight undulations, that are part of terraces or alluvial fans, and bush or shrub steppe as the dominant vegetation. The vegetation consists of subshrubs, scrub or cushions that grow despite poor soil conditions and severe weather. The dominant species include neneo (Mulinum spinosum), several species of senecio (Senecio sp.), colapiche (Nassauvia glomerulosa), neneo enano (Mulinum microphyllum), solupe (Ephedra frustillata), abrojo (Acaena platyacantha), yerba loca (Tetraglochin alatum), cactus (Maihuenia patagonica) and leña de piedra (Azorella monantha). The shrub species include quilimbai (Chuquiraga avellanedae), yaoyín (Lycium chilense), molle (Schinus johnstonii), calafate (Berberis microphylla) and mamuel choique (Adesmia volckmannii). All of them are generally represented by small and isolated specimens (Fig. 2). There are also landscape-defining species; for instance, the grassy steppe is dominated by coirón amargo (Stipa speciosa, S. humilis), coirón blanco (Festuca pallescens), coirón poa (Poa ligularis), cebadilla patagónica (Bromus setifolius) and cebada patagonica (Hordeum comosum), whereas a significant proportion of bare soil (up to 70%) has been noted (Fig. 3). The presence of such a diverse flora throughout Patagonia suggests the current environmental conditions in the region are not coincident with the conditions necessary for the production, transport and accumulation of sediments comparable with the ‘Rodados Patagónicos’. In fact, the present-day climatic and geomorphological conditions are far from meeting the requirements necessary for the generation of the Patagonian Shingle formations.

Figure 2.

Shrub or subshrub steppe with herbaceous and woody species, dominated by Asteraceae.

Figure 3.

Landscape units represented by a grass steppe, dominated by Poaceae.

ABOUT THE MORPHOGENESIS OF GRAVEL MANTLES

The extreme roundness that characterizes the gravel that forms the ‘Rodados Patagónicos’ (Fig. 4) suggests they were weathered by surface water transport, i.e. fluvial environments, using the term in its broadest possible sense. Other lines of evidence also support their fluvial origin. Firstly, remains of ancient drainage networks can often be identified on the surface of these plains using satellite images, aerial photographs and even in the field (O. A. Martinez, pers. observ.). Such ancient drainages can be related to at least the superficial gravel beds. Secondly, the field morphology of some of these deposits can occasionally provide a genetic link with surface water run-off, especially when they preserve the shape of alluvial fans, ‘bajadas’ (alluvial fans that are laterally coalescent) and pediments (erosion and transport surfaces, of very low slope, which are also generated at the expense of retreating mountain slopes). However, although such morphological features are highly diagnostic, they are only visible in younger units of the ‘Rodados’ and are not recognizable in older units. Thus, there are cases where the morphology does not provide the necessary clues for the assessment of the origin of these gravel mantles. These usually involve the most extensive gravel mantles, those covering hundreds of kilometres in a latitudinal sense and exhibiting a physical continuity between the Andean mountains and the Atlantic Ocean. They are a serious challenge for the reconstruction of their origin.

Figure 4.

Unit of ‘Rodados Patagónicos’ in northern Patagonia.

Two types of water flows can be identified that are capable of generating the ‘Rodados Patagónicos’ acting from the time the rock fragments were extracted from their source to the time when they were finally deposited: (1) intensive pulses, produced in a relatively short time under a high-energy regime; or (2) extended time regimes under conditions of low to moderate energy. Large amounts of ‘Rodados Patagónicos’ were deposited during relatively long periods and therefore both water regimes likely took part in their genesis. The energy of water flows is defined essentially by two variables: flow and land slope. Flow is a function of the climate (rainfall amount and/or melting) and land slope is a function of vertical ascents and descents of the land by tectonics (vertical movements on the margins of tectonic plates), epeirogenesis (vertical movements of larger continental blocks) and eustatism (changes in sea level). It is possible for gravel units exhibiting similar characteristics (size, thickness, sedimentological properties) to have been deposited at different times as a result of distinct processes during non-overlapping periods of sedimentation of different extent. As such, it is necessary to explain the origin of the gravel mantles of the ‘Rodados Patagónicos’ considering the Late Cenozoic, a time when hydrological and environmental conditions were notably different from today. In any case, whatever the origin of the gravel units under consideration, their origin can be linked directly or indirectly to climatic events that characterized the glaciations that began in the Late Miocene. Four scenarios considered most favourable for the generation of these sedimentary units are described in Table 1. These scenarios correspond to: (1) full-glacial periods; (2) fluvial extra-Andean piedmont environments during glaciations; (3) glacial ‘terminations’; and (4) fluvial piedmont environments (both Andean and extra-Andean) during interglacial periods.

Table 1.  Scenarios and environmental conditions considered most favourable for the generation of ‘Rodados Patagónicos’
EventGenesis of the ‘Rodados Patagónicos’Event durationSea levelClimateHydraulic gradientVegetation
AndeanExtra-AndeanAndeanExtra-Andean
GlaciationsGlacifluvial8000–15 000 yearsLoweredGlacial (very cold and wet)Periglacial (very cold and dry)HighNoSteppe/tundra
Piedmont
InterglacialsPiedmont10 000 yearsHigherSimilar to the presentSimilar to the presentVariableSimilar to the presentSimilar to the present
Terminations (transitions)Glacifluvial/fluvialVery short (centuries?)RisingSimilar to the presentSimilar to the presentDecreasingRecolonizationRecolonization

Accumulation of‘Rodados Patagónicosduring glacial periods

The analysis of the Last Glacial Maximum (LGM, Marine Oxygen Isotope Stage 2), the most recent and best-known glacial stage, allows a general characterization for the purpose of this study that could, to some extent and with great caution, be extrapolated to previous glacial events. This glaciation was a global climate event involving a decrease in the mean annual atmospheric temperature with respect to current temperature of approximately 6° in northern Patagonia and even more southwards (Clapperton, 1993; Rabassa, 2008). In Patagonia, this temperature decrease would have been accompanied by an increase in precipitation (Villagrán, Moreno & Villa, 1995), probably attributable to a northward migration of the westerlies (Markgraf, 1993; Bradbury et al., 2001; Heusser, 2003). During this climatic event, a mountain ice sheet > 2 km thick occupied the entire Patagonian Andes (from central Neuquén Province to Tierra del Fuego, see Clapperton, 1993). At its southern end, this continuous ice mass exceeded the present Atlantic coast, while slimming and vanishing around 37°S, from where the glacial accumulations consisted in cirque and valley glaciers located near the top of the mountains (Rabassa, Coronato & Martínez, 2011). The extension of the ice mantle from the Andean mountains to the east was accompanied by the development of periglacial environmental conditions (cold and dry) on the expanded Patagonian steppe, the Atlantic coast of which was displaced by between 200 and 300 km to the east with respect to its current position, as a result of lowering sea level (Ponce et al., 2011). Evidence of cryoturbation in ‘Rodados Patagónicos’, associated with permafrost development, has been noted for northern Patagonia at this time (Trombotto Liaudat, 2008). Under such conditions (which may have been representative of Late Cenozoic glacial events), processes leading to the accumulation of ‘Rodados Patagónicos’ would have been favoured, firstly, during the full glacial stages and, secondly, in extra-Andean fluvial piedmont environments.

Glaciofluvial ‘Rodados Patagónicos’ deposited in full-glacial stages

The vast majority of glaciofluvial deposits are generated during the stages of glacier stabilization. This stabilization occurs when ice masses come into equilibrium with the prevailing climatic conditions. From a glaciological perspective, it involves the setting of the margins (the glacier does not grow or diminish in size), with the amount of ice incorporated by snowfall and new generation being equal to the amount loss by melting (mass balance = 0). Under these conditions, in the terminal area located downstream where ablation is higher, two morphogenetic events dominate. Firstly, the debris in contact with the ice margin is incorporated upstream and is accumulated within the glacial mass. These materials, of varied shape and grain size, form ridges called marginal moraines (frontal and lateral). When preserved over time, these moraines indicate the position of the glacier margin after it has melted. Secondly, also in the terminal area of the glaciers, but affecting a large area downstream, important waterways are born from the base or sides of the glacier, which carry away from the moraines the brownish glaciofluvial sediments that build pro-glacial plains (washout) either from its base or sides. These plains are thus formed both by the debris contained in the ice and by the sediments removed from the moraines. The glaciofluvial flows tend to be very plentiful (depending on the size of the glacier and the melting rate), especially during the warmer seasons, and appear forming transition fans (alluvial) that can reach large sizes and will overlap laterally with time. Water courses usually have a braided pattern that changes constantly, typical of sediment-saturated currents (Fig. 5). These sediments are gradually finer and more rounded as they reach more distant positions with respect to their parental marginal moraines. As mentioned above, grain size, roundness and sphericity of the gravels are directly related to water transport capacity and distance transported.

Figure 5.

Present glaciofluvial streams (braided rivers), similar to the ancient glaciofluvial systems during the ‘terminations’ or full glacial phases.

In eastern Patagonia a large number of ‘Rodados Patagónicos’ plains can be linked genetically, with reasonable certainty, to former glaciofluvial processes. This is the case of deposits that are spatially connected upstream (toward the west) with marginal moraines. Several of these ‘rodados’ units, especially those located at higher latitudes, extend eastward, often reaching the present Atlantic coast. These outstanding deposits were generated under very rigorous environmental conditions, with a glacial climate in the vicinity of the Andes and periglacial conditions in the extensive extra-Andean area. Thus, the courses of sediment-laden waters, which came from the ice fronts, circulated during the winter over land that tended to freeze (permafrost), contributing to their own total or partial freezing. Such a landscape is thus likely to have strongly restricted the development of vegetation and associated wildlife, especially aquatic. However, during the winters, the migration of terrestrial wildlife would have been favoured in a north–south direction, probably on a regional scale. During the summers, the higher regional temperatures and the consequent increase in the melting of water from the Andean glaciers significantly modified the dynamics of extra-Andean ecosystems. Thus, the development of vegetation dominated by Poaceae and other elements typical of the steppe, particularly species of Asteraceae (Mancini et al., 2008), would have been favoured in those areas that escaped the influence of the mighty glaciofluvial currents. These same streams, in many cases saturated with sediment, would have limited the north–south migration of terrestrial species (in summer), while at the same time facilitating the movement of aquatic species adapted to these conditions in all available directions, as appears to have been the case for the freshwater fish Percichthys trucha (Ruzzante et al., 2011).

Accumulation of gravels in extra-Andean piedmont environments during glaciations

The majority of the accumulations of ‘Rodados Patagónicos’ deposited during the glacial events are likely of glaciofluvial origin. Piedmont deposits developed in extra-Andean areas, however, also exist. These deposits have likely formed as a result of the adjustment of the drainage networks to the lowering of the Atlantic Ocean or tectonic pulses superimposed on the glaciation effect. They exist in the form of alluvial fans, bajadas and pediment cover of very varied lithological composition, extension and thickness. Such piedmont deposits are characteristic of arid and semiarid climates and accumulate at the foot of the mountains where there is a significant change in the topographic slope and where the streams flow seasonally with a torrential regime, and abandon the deep valleys typical of the mountain environment to enter more open areas of lower energy, usually depressions between the ranges.

In our view, another significant variable, indirectly linked to the formation of the gravel plains, was the production of debris by cryoclastism, a phenomenon characteristic of periglacial environments, and it would have strongly affected, during each of the glaciations, much of the rocky outcrops that form the extra-Andean highlands. Much of these angular materials, of varied grain size and composition, were released from bedrock by physical weathering and must have been rapidly incorporated into the river networks when climate changed and rainfall increased in the area, with the consequent increase in the availability of surface water. Under this scenario, it is not easy to establish whether the glacial periods favoured or limited the processes of piedmont sediment accumulation in extra-Andean areas. While local fluvial processes were greatly reduced during glacial periods, the intense disintegration of rock outcrops by cryoclastism (physical weathering) would have increased the availability of loose surface material, which, subsequently, during wet periods, would have been captured by the drainage network.

It should also be noted that tectonic pulses or epirogenetic uplift likely also contributed to the accumulation of gravels in extra-Andean regions. The longer timescale of these events of endogenous origin (usually millions of years) widely exceeds the average duration of glacial and interglacial stages (measured in tens of thousands of years), and their footprint on the Cenozoic fluvial processes of Patagonia, undoubtedly very important, was superimposed on these weather events.

Accumulation of‘Rodados Patagónicosduring Glacial terminations

This section should be considered as the development of a hypothesis in which the word ‘termination’ is treated in a very flexible way. It is defined as a relatively rapid transition from a cold stage (glacial) to a warmer one (interglacial). Most authors who refer to this issue (see Lowe & Walker, 1997) have approached the analysis of the so-called Terminations I and II, the best known at present. The first one corresponds to the transition from Isotopic Stage 2 (Last Glaciation Maximum) to Isotopic Stage 1 (the Holocene or current interglacial); the second is the transition from Isotopic Stage 6 (Penultimate Glaciation) to subestadial 5e (warmest climate of the interglacial corresponding to Isotopic Stage 5). These periods of profound climatic changes were characterized by alternating minor climatic pulses in the context of a general warming. Termination I included the Late Glacial, a period when glaciers started to vanish from the continents. Generally, the transition from a glacial to an interglacial stage is much faster than one from an interglacial to a new glacial stage (Clapperton, 1993; Lowe & Walker, 1997). The installation, then, of a warmer climate during any glacial termination would have meant the abrupt transformation (or passage?) of the Cordilleran ice masses (Lowell, Heusser & Andersen, 1996) to a non-balanced climatic phase dominated by ablation and the decrease in thickness and area covered by the glaciers until their almost complete disappearance. From the standpoint of water, this meant a huge and sudden meltwater production resulting in rivers with a permanent high water discharge régime. Assuming Patagonia was dominated at those times by a marked climatic seasonality similar to the present (dry summers, rainy winters), it is likely that rivers achieved their highest volumes during the driest season (summers). Other factors to be considered for a correct understanding and reconstruction of these particular periods are, firstly, the gradual but influential rise of sea level and, secondly, the development in some mountain valleys of immense proglacial lakes enclosed to the east by the marginal moraines and to the west by the receding glacier. Under this scenario, the extra-Andean river valleys, which concentrated glacial flows during the full glacial period, were hydrologically disadjusted and subjected to processes of widening, but also of deepening of the river bed. This vertical erosion can be the cause of the terracing that characterizes most of the gravel plains, but this phenomenon would have been gradually weakened, especially in the eastern fringe closest to the coast, by the gradual rise of sea level (the regional base level of all watersheds). It is very likely that the underfit conditions, with respect to the present flow régime, of many of the Patagonian valleys (the Chubut, Genoa, Deseado, Santa Cruz and Coig rivers, among others; Fig. 1) is attributable largely to the strong erosion effect of water during these fast transitions from a glacial to an interglacial stage. It is also likely that ‘rodados’ units removed from their older locations to the west began to accumulate in positions further east in response to the decreased transport capacity of the rivers. According to this hypothesis, some of the ‘rodados’ units located in easternmost Patagonia were likely formed during these terminations.

Accumulations of piedmont‘Rodadosduring interglacial periods

In Patagonia there are many areas of ‘rodados’, whose origin has been assigned by different authors to piedmont fluvial action. While the origin for some of these is yet in doubt, there are remnants of some gravel units, especially very old units located at higher topographic heights, which were unquestionably deposited by water flows in a piedmont environment (see Fidalgo & Riggi, 1965; Panza, 2002). As already mentioned, piedmont deposits and landforms are originated by fluvial processes in arid or semi-arid climates, in response to reactivation of the landscape. These ‘rejuvenations’ consist of modifications in the base level of drainage networks attributable, in most cases, to the action of a tectonic uplift/depression (as a result of the action of tectonic plates) or epirogenic (vertical movements of large crustal blocks). These phenomena of endogenous character, a product of lithospheric deformation (diastrophism), have in the past frequently affected both Andean and extra-Andean Patagonian grounds. Moreover, such movements or diastrophic phases occur during periods that are often measured in hundreds of thousands or even millions of years. These processes have overlapped temporally with the climatic events (glaciations and interglacials) of the Quaternary. It should thus be safe to assume that at least some of the piedmont ‘rodados’ units accumulated during time lapses that included several and diverse climatic events.

As it was already mentioned, the piedmont deposits are characteristic of dry climates with low rainfall. Thus, it may be assumed that (applying actualism) weather conditions such as those currently prevailing in much of extra-Andean Patagonia (with mean annual precipitation below 300 mm, with seasonal, but not torrential rainfall) do not offer serious limitations, and should be sufficient for the formation of ‘rodados’ units throughout.

However, beyond the fact that many valleys and sides of different mountains and plateaus distributed throughout Patagonia are currently being modified, covered and obliterated by the development of alluvial fans (which in many cases tend to coalesce and generate morphologies called ‘bajadas’), it is safe to assert that these modest piedmont aggradation processes are quite different, because of their scale, from those that correspond to any of the great ‘Rodados Patagónicos’ units that lie in the area.

If significant rainfall is not needed to justify the presence of these gravel mantles, it is then necessary to incorporate in their genetic scheme the influence of a topographic gradient sufficiently large to have allowed intensified water erosion in elevated areas, the consequent downstream transport of extracted materials and their accumulation, rounded, in distal positions of the piedmont. This increased hydraulic gradient, that is, the increase in energy availability in the system, was provided by diastrophic movements in Patagonia since the early Neogene (approximately 20 Ma). These movements can be summarized as the Pehuénchicos movements that occurred at the Oligocene–Miocene boundary, the Quéhuchicos movements in the Late Miocene and the Andicos movements during the Pliocene and even in the Pleistocene (Ramos & Ghiglione, 2008).

FINAL REMARKS

The ‘Rodados Patagónicos’ are sedimentary units distinguished by their large areal extent, their topographic monotony and uniformity, and their marked sedimentological homogeneity, characteristics that keep a significant independence with the actual age of the deposit. Currently, either in Patagonia or any other place, processes that lead to the generation of equivalent or similar gravel accumulations do not exist, despite the great variety of climates and ecological conditions that presently exist on the Earth's surface. Patagonia has therefore likely experienced unique conditions in the recent geological past (at least from the late Miocene onwards), that were favourable for the production, transport and sedimentation of these characteristic materials on a large scale. Far from solving such an intriguing topic, we consider that, beyond the geological and climatic variables, the initial relief (essentially the topography and regional slope) on which the different units of rodados were deposited would have played a very important role. Assuming that solely the action of surface water runoff may explain these sedimentological and geomorphological properties characteristic of the ‘Rodados Patagónicos’, sufficiently long periods of the past need to be identified with fluvial processes of the magnitude and intensity necessary for these accumulations to have taken place.

The glacial stabilization (full-glacial phase) and glacial terminations (a term used in a broad sense to indicate any fast deglaciation) bring together, for different reasons, the most favourable conditions for the development of many of the Patagonian gravel units, both in terms of their sedimentology, as in the occurrence of landforms such as plateaus and terraces. The environmental conditions under which these materials were deposited are significantly different from each other. The extra-Andean Patagonia vegetation was changing in response to these different environmental conditions. While, in general, the physiognomy corresponded to grass and subshrub/shrub steppes, and tundra during glacial periods, their specific composition and coverage were changing, and therefore the associated fauna did so as well.

In periods of high input of water, with permanent and high discharge streams, the continuous generation and deposition of sediments and ‘rodados’ in the flood plains would have modified seasonally the vegetation cover, represented mainly by annual herb species and aquatic plants (macrophytes) associated with water courses, which year after year restarted a new process of colonization, the plant succession. The higher water supply would have facilitated the dispersal of some aquatic species (e.g. see Ruzzante et al., 2011), whereas for terrestrial species it may have acted as geographical barriers to migration, thus limiting the north–south migratory processes.

In contrast, certain gravel mantles, some of them very extensive, located on the eastern strip of Patagonia (near the Atlantic coast), but also other relicts found in the mountains, seem to be better explained by strictly fluvial piedmont processes. Those which responded to tectonic/epirogenic reactivations had their maximum activity during non-glacial periods, when the periglacial conditions were not yet developed in the extra-Andean area.

Although the terms uniformity and homogeneity appear to be an unavoidable part of any definition about the ‘Rodados Patagónicos’, evidence suggests that, paradoxically, their genesis would be linked to a variety of environments and ecosystems. Not only they would have been deposited at different times of the Late Cenozoic, under very different climate types, but also the same unit of rodados, especially those born in the Andean area and ending along the Atlantic coast, were generated by streams that passed through longitudinal strips with very different ecological characteristics.

ACKNOWLEDGEMENTS

We thank Jorge Rabassa, Eduardo Tonni, Alfredo Carlini and Daniel Ruzzante for the invitation to participate in the Symposium ‘Palaeogeography and Palaeoclimatology of Patagonia: Effects on Biodiversity’, held at the La Plata Museum in Argentina in May 2009.

Ancillary