IRDDive

Mining documents with IRD data over the last 5 millions years.

By: Jeremiah Marsicek

Date: 02/07/2018

Purpose:

To get information about the spatiotemporal extent of IRD events over the last 5 million. The ultimate goal could be to make a series of maps with the publication date, publication, and latitude/longitude and timing of IRD events.

Main Resources:

We will use R and RStudio to do the data processing, GeoDeepDive to retreive documents of interest via string matching and to generate a subset of the output (for testing) as NLP output.

What are some example publications?

1. Ice-rafted debris associated with binge/purge oscillations of the Laurentide Ice Sheet (http://onlinelibrary.wiley.com/doi/10.1029/94PA01008/full)

2. Catastrophic ice shelf breakup as the source of Heinrich event icebergs (http://onlinelibrary.wiley.com/doi/10.1029/2003PA000890/full)

3. The sources of the glacial IRD in the NW Iberian Continental Margin over the last 40 ka (https://doi.org/10.1016/j.quaint.2013.08.026)

Using these example publications I have found the these terms appear the most:

Terms for IRD: iceberg rafting, IRD, ice-rafted debris, Heinrich events

These papers talk about IRD in the following locations: North Atlantic, Ruddiman Belt, Hudson Bay, Antarctic

Most common time frames: last glaciation, Pleistocene, Holocene, year subsets

Targets with NLP output from GDD team:

Extract latitude and longitudes using Simon's code as a template (which uses regular expressions to get lat/long).

Place names using “Who’s on First” (a large Postgres database of place names with associated polygons).

Extract time periods using regular expressions and queries for stratigraphic names and regular expressions.

Accessing the RMarkdown document as an html:

http://htmlpreview.github.io/?https://github.com/EarthCubeGeochron/IRDDive/blob/master/empty-spaces.html