Academia

Doing Bayesian Lead-210 interpretation

Paleolimnologists have been using the Constant Rate of Supply (CRS) and Constant Initial Concentration (CIC) model to interpret 210Pb dates for a very long time. Some variations like propagating error using a Monte Carlo simulation (Binford 1990 and Sanchez-Cabeza et al. 2014) instead of the usual “quadrature” method have been used, but for the most part, we have been interpreting lead-210 dates the same way since 1978 (40 years!).

Reconstructing a coagulant optimization model using R and Shiny

Inspired by Lindsay Anderson’s suggestion that “somebody code up that Edwards 1997 model”, I successfully avoided proofreading my Ph.D. thesis proposal for much of a day doing just that. The paper is a classic paper, and one of many that attempt to predict dissolved organic carbon removal using easily measurable water quality parameters (in this case, absorbance of UV radiation at 254 nm, influent DOC concentration, and coagulant dose). Lindsay has a great paper talking about why coagulant dosing is important in an era of rapidly changing source water quality!

The Circumpolar Diatom Database using R, the tidyverse, and mudata2

It is an exciting time for the integration of limnological and paleolimnological datasets. The National (US) Water Quality Monitoring Council Water Quality Portal has just made decades of state and federal water quality measurements available, the Pages2k project has collected hundreds of temperature proxy records for the last 2000 (ish) years, and the Neotoma database provides access to a large number of paleoecological datasets. For a final project in a course last fall, I chose to analyze the Circumpolar Diatom Database (CDD), which is a collection of water chemistry and diatom assemblage data hosted by the Aquatic Paleoecology Laboratory at ULaval.

Comparing approaches to age-depth modelling in R

Working with radiocarbon dating in R has long been possible, especially since the Intcal dataset itself contains R code in the supplement. Other tools like Bacon (Blauuw and Christen 2011), the slightly simpler Clam (Blaauw 2010), and BChron (Haslett and Parnell 2008) have helped users calibrate radiocarbon dates and produce reproducible age models. A comprehensive analysis of the quality of these and other methods is available in an article by Trachsel and Telford (2016).

Abbreviating journal titles using BibTex & R

It seems that the tools for writing papers in R/RStudio keep getting better and better, to the point where it is rare that I have something I need to do to write a paper that happens outside of RStudio. One of these things is abbreviating journal names, because for whatever reason the checkbox that does this within Zotero’s BibTex export doesn’t work particularly well. My way around this in the past was to wait until the article was about to be submited, and figure everything out in Microsoft Word at the very end.

10,000 Years of Metal Dynamics in the Tantramar Marshes

Almost six years ago, my honours thesis (Using paleolimnological methods to track late holocene environmental change at Long Lake, New Brunswick – Nova Scotia border region, Canada) and my friend Hilary’s master’s thesis were completed, but it was only this year that we finally got around to submitting the combined manuscript to FACETS, a new (ish) open access journal from Canadian Science Publishing. Hilary’s thesis describes (among other things) a peak in mercury deposition at Long Lake, near Amherst, Nova Scotia, approximately 5,000 years ago.

Modeling sediment deformation using R & RMarkdown

Last summer I came home from a hike one day with way too much thesis work to do (my submission date was, I think, close to a week away). As per usual, I stopped by the K.C. Irving Centre to catch up on work, and promptly sat down and started writing what would become my latest publication in the Journal of Paleolimnology. The article used R to apply mathematical functions describing how sediment is deformed when cored, and for the first time in my short career of academic writing, I wrote the paper in RMarkdown.

New article!

It’s been a nearly 9-month saga, but I’m pleased to announce that my first ever peer-reviewed paper has been published in the Journal of Paleolimnology as of October 2016. The paper (”A geochemical perspective on the impact of development at Alta Lake, British Columbia, Canada”) is a result of my M.Sc. work on Alta Lake in Whistler, British Columbia. Hopefully the first of many to come!