Data visualization: R color palettes

Here are some of my favorite R packages for color palettes:

  1. “pals”, which has a nice set of examples shown here: https://cran.r-project.org/web/packages/pals/vignettes/pals_examples.html

Pals essentially collects together in one place a lot of the classic color palettes, and gives you some new tools to explore them. It contains all the “viridis” color schemes, as well as some palettes I’ve seen more frequently in matlab such as “parula” (I really like parula, and used it our paper here). I also really like “coolwarm” as a diverging palette and used it here.

2.  @aljrico has developed some really nice color palettes for R based on Harry Potter or Game of Thrones. I really like the Harry Potter ones, particularly Ron Weasley (yes, it has some orange) and Luna Lovegood! See: https://github.com/aljrico/harrypotter

I used Ron and Luna in my recent preprint here.

R.E. Baker the beer…

My most surreal moment from 2020 (pandemics aside) was getting an out-of-the-blue email saying that a beer had been named after me. Sometimes I like to look at reviews for R.E. Baker the hazy IPA, on Untappd. The beer cites my climate/covid paper, though unfortunately google scholar doesn’t count beer citations…

https://untappd.com/b/postdoc-brewing-company-pop-quiz-r-e-baker/3964425

Dynamic response of airborne infections to climate change: predictions for varicella

I have a new paper out in Climatic Change, co-authored with the brilliant Ayesha Mahmud and Jessica Metcalf. The first figure in the paper is this striking correlation between state-level estimates of varicella R0 (a measure of disease transmission potential) and average humidity for those states. In the paper we do a lot of work to move from this high-level observed correlation, to a richer understanding of the climate drivers of varicella, disentangling other factors that might bias this observed relationship. In the process we attempt to synthesize mechanistic approaches that capture the dynamics of these types of diseases very well, with statistical approaches required to tease out climate effects from observational data records. Nevertheless, this macro-level correlation still stands to motivate our intuition: our main result (after much more modeling work) supports this observed relationship.

The cool thing about our method is we end up being able to generate future simulations that capture the known dynamics of the disease and incorporate climate dependence. When we force these simulations with projections for changing environmental conditions, we find that climate change could cause a seasonal shift in cases from winter to summer months. This finding, that climate change may interact with the dynamics of airborne diseases, may have implications for public health provision. It is something we hope to investigate further in future research!

Check out the full paper here! https://rdcu.be/MYzM