Date of Award

December 2016

Degree Type


Embargo Date


Degree Name

Doctor of Philosophy (PhD)


Earth Sciences


Gregory D. Hoke


Amazonia, Andes, cosmogenic radionuclides, erosion rates, landscape evolution, paleo-erosion rates

Subject Categories

Physical Sciences and Mathematics


In the past three decades, the use of cosmogenic radionuclides revolutionized research in geomorphology. Cosmogenic radionuclides, such as 10Be and 26Al, accumulate within quartz crystals at a relatively well-constrained rate due to the bombardment of the Earth’s atmosphere by cosmic rays. Since the early 90’s, many studies utilized 10Be and 26Al to determine the pace and timing of vertical changes, or absence thereof, to the uppermost meters of the Earth’s surface. In particular, the ability to simply collect a bag of sand from the bottom of a stream and, after extracting the quartz from it, determine the average rate of surface lowering (i.e. the erosion rate) of the integrated catchment area upstream of that point, allowed many researchers to explore the linkages between climate, tectonics, and surface processes in evolving landscapes. In contrast to dozens of studies investigating the modern erosion rates, less than ten studies actually pushed the boundaries of this application to seek paleo-erosion rates by querying older sedimentary rocks for their 10Be content from the time the sediments were deposited. Analogous to the modern case, in which the cosmogenic nuclide inventory of alluvium reflects erosion rates, the cosmogenic nuclide inventory of quartz in independently well-dated sedimentary rocks contain a decayed dose that can be converted to a paleo-erosion rate. The work outlined within this dissertation builds on these concepts and combines the concentrations of radionuclides from modern river samples as well as sedimentary rocks to determine the pace and timing of landscape change in the Central Andes of Argentina and Central Amazonia.

To date, the limits and assumptions embedded in the use of paleo-erosion rates have only been lightly addressed in the literature. No tool exists for the assessment of the sensitivity of paleo-erosion rate data to other site-specific variables, such as past sediment accumulation rates, unconstrained pre-sediment burial exposure time, and exhumation of previously buried deposits. Chapter one describes the well-established concepts behind the accumulation of 10Be and 26Al in quartz in different scenarios in which a parcel of rock or sediment exhumes or descends towards or away, respectively, from the Earth’s surface, and provides a tool that can help researchers identify the limits of the applicability of this method for site-specific conditions.

Chapter two consists of the application of the methods described in Chapter one to build a record of paleo-erosion rates from the late Miocene to the early Pleistocene by probing the sedimentary rocks from the foreland of the Argentine Andes at 30°S. Here, approximately 5.5 km of sediment accumulated as the result of erosion of the adjacent Andes from ca. 15-2 Ma. The strata in the foreland were previously dated through magneto-stratigraphy, which provide independent age and burial rate constraint of the targeted units. Also, the history of deformation of the Precordillera fold-and-thrust belt is well constrained from both dating of thrust-top synsedimentary deposits and 2D balancing of cross-sections, which were confirmed through (U-Th-Sm)/He thermochronometry. Together, the knowledge established in the literature and the paleo-erosion rate record built in this study allowed the comparison of the histories of deformation, erosion, and sediment accumulation. With paleo-erosion rate records built for one section in the wedge-top, upstream of the Precordillera, and two from the foredeep, downstream of the Precordillera, the dynamics of erosion in concert with major deformation and mountain building in the Precordillera were constrained. The findings revealed a >2 Ma lag between the horizontal shortening rate (12-9 Ma) and erosion rate (7 Ma) maxima. This result has implications for our understanding of landscape evolution and mass-flux during orogenesis, especially critically tapered ones. Particularly, it can better inform modeling efforts that focus on tectonic and climatic influences on those mass fluxes.Chapter three consists of a synoptic view of the modern pattern of erosion in the arid South-Central Andes between 28°S and 36°S. Here, new millennial and decennial erosion rate data were produced for nine catchments draining the high Andes from the Argentine side as well as the Precordillera. These new data were combined with published data from arid catchments in Argentina both from the vicinities of the study area and from the southern edge of the Altiplano-Puna plateau, all tectonically active and arid areas. The Argentine dataset was used for comparison with the along-strike pattern of millennial and decennial erosion rates published for the Chilean side, which experiences a steep precipitation gradient, going from arid to temperate climate in less than a latitudinal degree near 32°S. Together, this synoptic view of along-strike erosion rates revealed a nearly identical pattern of erosion rates on both sides of the Andes; both show enhanced erosion rates between 32°S and 34°S, which taper off southward. The erosion rates show no correlation with other parameters, such as precipitation and vegetation, and are closely, but nonlinearly correlated with the mean catchment local gradient. Interestingly, the along and across-strike patterns of erosion of the South-Central Andes closely follow a published southward increase in total surface uplift since 10 Ma. The similarity of erosion along-strike on both sides in spite of different climatic conditions indicates that erosion of this area is closely related to tectonics, and not climate as previously proposed.

Chapter four focuses on landscape changes in Central Amazonia, particularly in the lower Negro River valley and the Negro-Solimões interfluve. This region contains the topographic features of a completely abandoned, southward-flowing paleo-network that has been incised by modern northward-flowing rivers. It is of particular importance to the Amazonian landscape evolution not only because of its location near the junction of two major Amazonian rivers, the Negro and Solimões Rivers, but also because past environmental changes in the Negro River

valley promoted faunal endemism in the last 1 Ma. Using cosmogenic 10Be and 26Al, a new late Miocene burial age is constrained for the paleo-channel deposits, indicating that this landscape is older than previously thought, bearing important implications for the landscape and species evolution. Geomorphic analysis of fluvial metrics derived from a digital elevation model revealed a signal of base level fall in the Negro River valley, consistent with a modification of the valley or the Negro River channel itself. Given the timing of burial and the existence of a ubiquitous base level fall signal in the region, the paleo-network is probably a remnant of the ancestral Amazon drainage system. Its deflection was potentially facilitated by faulting in the lower Negro River valley and may be related to the regional tilting associated with the formation of the transcontinental Amazon River system caused either by dynamic topography or progressive flexural loading of the upper crust from west to east.


Open Access

Available for download on Saturday, July 28, 2018