LinkedEarth: what is it, what it is not.

We have a lot of ideas for LinkedEarth.  Since not everyone lives in our head, in this post I'll lay out in plain terms what this project is about. 

LinkedEarth is:

  • a platform to crowd-curate paleoclimate data. Our experience with the PAGES2k project is that the lack of such cyberinfrastructure has wasted thousands of (wo)man-hours, and that future efforts at data synthesis will be greatly aided by having a simple platform on which to upload, edit, query and download paleoclimate data. 
  • an effort to standardize paleoclimate data. This includes metadata, so it's easy to search records because they're all described the same way.  It also aims to standardize the reporting of certain type of data. For instance, except in rare cases, there does not appear to be a community standard on how to report radiometric age constraints, be they from [latex] {}^{14}C , U/Th, {}^{210}Pb, {}^{137}Cs[/latex] or chronostratigraphic markers (e.g. volcanic eruptions of known age). This is a problem both within communities (e.g. the radiocarbon community) and for scientists who rely on multiple age constraints to form their age models.  We're hoping that LinkedEarth standardization efforts will spur these various communities to come together and find ways to represent radiometric ages in common ways.
  • an incubator for cloud-based paleoclimate data analysis.  Though it isn't part of our current mission, we'd very much like the LinkedEarth platform to enable direct visualization of paleodata, as well as the results of some super useful analyses, like age-uncertain calibrations, age-uncertain spectral analysis, age-uncertain phase relations between records, comparisons between age-uncertain records, etc. See a common theme here? Age Uncertainties!  This is highly synergistic with the GeoChronR project (PI: McKay, co-PI: Emile-Geay), and though it's unlikely that we'll be able to fully integrate them in this current funding cycle, we're very much hoping to do so in the next one.  
  • a social experiment.  What happens when you ask dozens of people with different expertise to come together and agree on a common vocabulary? You're about to find out!  Seriously, we are optimistic about this because (as is often the case), paleo is late to the party. Other scientific communities have gone through this before and formed controlled  vocabularies. The Gene Ontology is an example of the spectacular results that ensue when communities come together to do such things. 

 LinkedEarth is NOT:

  • the answer to all your troubles
  • a free customer service about quantitative data analysis
  • a substitute for other data repositories like NCEI Paleo or Pangaea. 
  • a DOI-minting organization 
  • a venting space about EarthCube

In short, we intend for LinkedEarth to become a one-stop shop to upload a new paleoclimate dataset, compare it to relevant records, perform some sophisticated algorithms to date and calibrate it, and graph that in intuitive ways.  Actually, we're thinking a lot bigger, but we don't  have the funding for that now. We are thinking not just paleoclimate, but all kinds of other paleo (paleoecology, paleobiology, archeology). We're thinking integration with instrumental datasets and model output. We're thinking, in a word, a full integration with the rest of EarthCube, assuming that other EarthCube projects deliver at the same level that we are.  

If you are excited about that vision,  if you want for EarthCube to deliver better ways to store, exchange and analyze geoscientific information, hop on our wiki. Give it a spin. Start discussing vocabularies with other scientists. This project is crowd-based, and it will only work for you if you get involved. It will be more fun for us that way.   And it will show that this can be a viable model for other geoscientific communities, so one day LinkedEarth actually does link together all manner of Earth Science data, not just paleoclimate data. We're quite busy as things stand, so we'll first try to do this well for paleoclimate - with your help of course. 


Leave a Reply