Most of these publications may also be found on my google scholar profile.
If you like the format of the preprints included here, see preprint-template.tex

Neoproterozoic glacial origin of the Great Unconformity

C. Brenhin Keller, Jon M. Husson, Ross N. Mitchell, William F. Bottke, Thomas M. Gernon, Patrick Boehnke, Elizabeth A. Bell, Nicholas L. Swanson-Hysell, and Shanan E. Peters
Proceedings of the National Academy of Sciences, 2018:
Preprint     PDF     Code and Data   

It has long been observed that the sequence of sedimentary rocks deposited in the past half-billion years often sharply overlies older igneous or metamorphic basement at an erosional surface known as the Great Unconformity. We provide evidence that this unconformity may record rapid erosion during Neoproterozoic “snowball Earth” glaciations. We show that the extent of Phanerozoic sedimentation in shallow continental seas can be accurately reproduced by modeling the accommodation space produced by the proposed glacial erosion, underlining the importance of glaciation as a means for lowering erosional base level. These results provide constraints on the sedimentary and geochemical environment in which the first multicellular animals evolved and diversified in the “Cambrian explosion” following the unconformity.

Press coverage:
National Geographic , Ars Technica , LA Times , , Live Science , How Stuff Works

Plate tectonics and continental basaltic geochemistry throughout Earth history

Brenhin Keller and Blair Schoene.
Earth and Planetary Science Letters, 2018:
Preprint     Code and Data   

In this longer-format paper we examine in detail the record of basaltic rocks preserved in the continental crust, building on the weighted bootstrap resampling procedure of Keller and Schoene, 2012. A range of rapid trace element variations in the basaltic record can be explained as a result of mantle melting systematics (changing partition coefficients as a result of Grt and Cpx-out). Both trace element signatures of slab fluid input and major element signatures of calc-alkaline vs tholeiitic differentiation are remarkably stable and consistently arc-like, strongly suggesting the occurence of subduction and plate tectonics throughout the preserved rock record (back to at least 3.85 Ga). While some non-plate tectonic models may produce some flux melting, a virtually constant proportion of flux to decompression melting in the preserved continental record is a tall order for any non-plate model for Archean tectonics.

Temporal variation in relative zircon abundance throughout Earth history

C. Brenhin Keller, Patrick Boehnke, and Blair Schoene
Geochemical Perspectives Letters, 2017:
Preprint     PDF     Code and Data   

Here we investigated the variation in the amount of zircon expected to saturate from magmas of different ages throughout Earth history by running full MELTS simulations along with trace element partitioning calculations (on a local Princeton University cluster) for each of the ~70,000 samples in the Keller & Schoene (2012) dataset. We found that older magmas crystallize substantially less zircon per unit mass due to the geochemical consequences of secular mantle cooling – suggesting that a larger volume of felsic magmatism is required to explain the Archean (and perhaps the Hadean) zircon record than previously considered. We note also that anorthosite flotation crust is largely zircon free and thus invisible to zircon-based crustal growth models, with implications explored in the supplement.

Volcanic–plutonic parity and the differentiation of the continental crust

C. Brenhin Keller, Blair Schoene, Melanie Barboni, Kyle M. Samperton, and Jon M. Husson.
Nature, 2015:
Preprint     Code and Data   

Here we used the weighted bootstrap resampling approach from our 2012 paper to address the longstanding question of volcanic-plutonic parity. The results indicate that felsic plutons are not (on average) significantly more cumulate than felsic volcanics, favor fractional crystallization as the predominant mechanism of geochemical differentiation, and suggest the influence of magmatic water content on magma stalling and intrusion. This paper was my first foray into high-performance computing, running ~1.3 million pMELTS simulations to invert for P-T paths and water contents that produce differentiation trends similar to the those observed in the natural dataset.

Statistical geochemistry reveals disruption in secular lithospheric evolution about 2.5 Gyr ago

C. Brenhin Keller and Blair Schoene.
Nature, 2012:
Preprint     Code and Data   

My first significant computational work - originally a side project to work on as a grad student while waiting for Blair's clean lab to be built. We were surprised by the strength and consistency of the geochemical trends we observed on Gyr timescales. The simplest signals directly reflect secular mantle cooling, but others hint at a change in crustal evolution that appears to coincide temporally with oxidation of the surface Earth in the Great Oxygenation Event.