Abstracts for Virtual Thermo2020

April 27, 2021, Session 7

Talk 1

Late Miocene–Pliocene onset of fluvial incision of the Cauca River Canyon in the Northern Andes

N. Pérez-Consuegra1*, G. D. Hoke1, P. Fitzgerald1, A. Mora2, C. Montes3, E. R. Sobel4, J. Glodny5

1 Department of Earth and Environmental Sciences, Syracuse University, Syracuse, NY, USA
2 Ecopetrol Brasil, Rio de Janeiro, Brazil
3 Departamento de Física y Geociencias, Universidad del Norte, Barranquilla, Colombia
4 Institut für Geowissenschaften, Universität Potsdam, Potsdam-Golm, Germany
5 GFZ German Research Centre for Geosciences, Potsdam, Germany

* nperezco@syr.edu

The incision of kilometer-scale canyons into high-standing topography is often used to constrain the surface uplift history of mountain ranges, controlled by tectonic and geodynamic processes. However, changes in climate, such as orographic enhancement of precipitation during mountain growth or periods of time with increased rates of precipitation, may also be responsible for canyon incision. This study deciphers the timing and the role of tectonic/climatic mechanisms on the incision of the ~2.5 km deep Cauca River Canyon in the Central Cordillera of the Northern Andes using the cooling (exhumation) history of rocks from the canyon walls. Ten bedrock samples and one detrital sample were collected on the eastern border of the canyon between 300 m and 2,300 m elevation. Bedrock and detrital AFT data yield ages from 38 to 50 Ma, while two bedrock AHe ages from the valley bottom yield ages of 6–7 Ma. The AHe cooling ages and inverse thermal history models reveal a previously unidentified Late Miocene (ca. 6–7 Ma) pulse of exhumation which corresponds to the onset of carving of the Cauca River Canyon. The absence of younger AFT ages in the Cauca River Canyon demonstrates that the magnitude exhumation during canyon incision was insufficient to reveal younger AFT cooling ages below an exhumed Partial Annealing Zone. The Cauca Canyon was carved because of increased rock uplift rates in the northern Central Cordillera and headward propagation of an erosion wave into the mountain range.

Talk 2

Coeval Miocene exhumation of the Cycladic Basement and the Cycladic Blueschist Unit in the southern Cyclades (Ios and Sikinos Islands, Greece)

Megan E. Flansburg1*, Eirini M. Poulaki1, Daniel F. Stockli1, Konstantinos Soukis2

1 Department of Geological Sciences, Jackson School of Geosciences, University of Texas at Austin, TX, USA
2 National and Kapodistrian University of Athens, Greece

* meflansburg@utexas.edu

The North Cycladic, West Cycladic, Naxos-Paros, and Santorini detachment systems accommodated large-scale Oligo-Miocene exhumation and formation of metamorphic core complexes in the backarc of the retreating Hellenic subduction zone.  While these detachment systems are either contained within the Cycladic Blueschist Unit (CBU) or at the CBU-Upper Unit interface, the sheared contact between the CBU and the underlying Cycladic Basement (CB) in the southern Cyclades has been debated for over 30 years, due to the coexistence of top-N and top-S shear sense indicators and dispersed ages from previous thermochronology.  New zircon and apatite (U-Th)/He ages and HeFTy time-temperature pathways for the CB and CBU on Ios and Sikinos islands show that both units cooled through these low temperature regimes from ~14-6 Ma suggesting that the units were juxtaposed prior to and exhumed together during bivergent Miocene top-S (Santorini) and top-N (Naxos-Paros) detachment faulting not directly exposed on Ios or Sikinos islands.

Talk 3

Evaluating (U-Th-Sm)/He data variability of the Durango apatite standard using a Monte Carlo error modeling approach

Peter E. Martin1*, James R. Metcalf1, Rebecca M. Flowers1

1 Department of Geological Sciences, University of Colorado Boulder

* Peter.Martin-2@colorado.edu

(U-Th-Sm)/He data commonly exhibit intra-sample variability beyond that predicted by analytical uncertainty. Possible contributions to data dispersion include varying diffusion kinetics (e.g., grain size, radiation damage), alpha particle redistribution (implantation, parent nuclide zonation), and parent nuclide-rich inclusions. A complete understanding of intra-sample variability is reliant on quantification of its exact extent, which requires a careful accounting of analytical uncertainty in (U-Th-Sm)/He data. However, derivation of the uncertainty in (U-Th-Sm)/He dates is not straightforward owing to the fact that the helium age equation lacks an analytical solution. We therefore developed a Monte Carlo error propagation code to derive uncertainty in (U-Th-Sm)/He data, which serves the dual purpose of permitting highly accurate uncertainty quantification and direct comparison to analytical methods of uncertainty propagation. These analyses demonstrate that the Monte Carlo and analytical methods of deriving uncertainty are nearly indistinguishable. Uncertainty generated using the Monte Carlo method tends to be very slightly larger (a difference of ~0.1% in relative standard error) and exhibits very minor skew in the probability distribution. These differences are so slight that the more computationally efficient analytical error propagation methods are acceptable in nearly all cases.

            We apply these uncertainty derivation methods to archived Durango apatite data from the University of Colorado Thermochronology Research and Instrumentation Laboratory (CU TRaIL). This de facto standard exhibits few of the above physical characteristics that potentially cause intra-sample variability, and the CU TRaIL has a large amount of archived Durango (U-Th)/He results. We analyze a subset of these data (n=234) acquired in a three year period following acquisition of new instrumentation in September of 2017. We find that Durango apatite exhibits intra-sample variability (standard deviation of ±2.6 Ma) beyond that predicted from analytical uncertainty alone (±0.6 Ma, 1σ). However, the shape of the histogram of data is distinctly non-gaussian, suggesting that a kernel density estimate (KDE) is a more accurate descriptor of the true data distribution. The KDE has a 68% confidence interval of [+1.8, -2.2], implying a smaller degree of overdispersion and reflecting a shoulder to the low side of the peak. The KDE also includes relatively long tails, suggesting that the large standard deviation of the data is a result of extreme values. This overall shape is consistent with prior modeling in the literature describing the effects of parent nuclide zonation in Durango apatite, suggesting that such zonation is the cause of overdispersion in these data.