PaleoBooks: Science Examples for Cyberpaleoclimatology
Over the past decade, the LinkedEarth team has proposed and promoted standards for paleoclimate data and software. To date, we have 5898 standardized datasets that are ready for use in scientific studies using these open-source tools. The question becomes: how does one get started with this open paleoscience ecosystem?
While we have dutifully documented all of our software packages and data standards, such documentation tends to focus on simple concepts on well-behaved examples. Real-world science, unfortunately, is rarely well-behaved.
Enter PaleoBooks, a gallery of reproducible computational narratives (aka “notebooks”) using open source tools in Python (and soon, R). So far, these notebooks have been sourced from class projects, conference presentations, or scientific manuscripts, and leverage the Jupyter Book technology. The goals of PaleoBooks is to allow any scientist to copy/paste snippets of our hard-earned code and reuse them in their own work while promoting open science. For instance, looks at this water mass visualization:
Beautiful, isn’t it? This is the work of PhD Student Jordan Landers. The code is available in PaleoBooks and can be adjusted and reused for other purposes (and water masses). I wish I had that code available when I was writing my PhD dissertation. It probably would have saved me many hours and a few headaches.
This is one of the purposes of PaleoBooks: to share useful coding tricks that everyone at some point in their career wished they had. To avoid re-inventing the wheel and get further down the road to discovery.
The second purpose of the gallery is to promote open science. Can you think of a more reproducible code example than a Jupyter Book describing the detailed steps of an analysis alongside the code? I can’t. Well, I can think of a few more ways but this is, currently, as good as it gets. In fact, PaleoBooks contain an example of an entire scientific manuscript, with every figure reproduced in a separate Jupyter Notebook, weaved together into one such “PaleoBook”: earlier this year, PhD student Alexander James contributed the results of his latest manuscript, Detecting Paleoclimate Transitions With Laplacian Eigenmaps of Recurrence Matrices (LERM), published in Paleoceanography and Paleoclimatology.
So what’s next for PaleoBooks? In the next year, we plan to add more examples of open paleoscience and we can’t do it alone! Indeed, we want to represent the broadest possible cross-section of digital paleoclimate studies, so we need YOU to contribute your science. If you feel the call, please reach out to the LinkedEarth team and let’s discuss how to get you started! Think about it: you get to make your science more reproducible, more visible, and more open, helping future generations of geoscientists stand on your shoulders. If you are already using Python and Jupyter Notebook, we promise that it won’t take much of your time to turn it into your own PaleoBook!
D.K.
PaleoBooks is one component of the PaleoCube project, supported by the U.S. National Science Foundation.
