Lunar geochemical datasets on MoonDB

The Apollo missions returned 2200 samples comprising “382 kilograms (842 pounds) of lunar rocks, core samples, pebbles, sand and dust from the lunar surface” (quoted from ref). Since then, we’ve sliced, diced, dissolved, vaporized, irradiated, and applied just about every other analytical tool to investigate these samples. An incomplete list of geochemical investigative techniques used includes: electron microprobe, instrumental neutron activation analysis, various mass spectrometry techniques (multi-collector, laser ablation, thermal ionization, time-of-flight, secondary ion), X-ray fluorescence, and cathodoluminescence. The results of these studies are scattered across the past 46 years of published literature, in various formats and accessibility.

Most lunatics probably have a ragtag collection of spreadsheets incorporating personally vetted data from the literature. But it is easy to miss good papers, and time consuming to re-type poorly scanned copies. I have spent a fair bit of time hunting down literature data on lunar basalts, and these important datasets are not always easy to find or re-use in a useful way. To make this process easier, MoonDB aims to serve as an open access data repository for lunar geochemical and petrological data. It is led by Kerstin Lehnert, PI of the EarthChem/PetDB project.

MoonDB_screenshot

MoonDB (http://www.moondb.org/) is launching a program to restore and preserve lunar geochemical analytical data. The framework will be nearly identical to the Petrological Database (PetDB), where geochemical data from terrestrial studies are freely available. The framework is still under development, but it appears they have data from 500+ papers archived and awaiting release.

In addition to incorporating peer reviewed article data, MoonDB plans to serve as a “publishing” venue for otherwise unpublished datasets. Researchers occasionally collect more data then they eventually need to include in a manuscript. Or they might perform foray analyses on samples for projects that never go anywhere. MoonDB hopes to rescue these from eventual hard drive failure and publish them in a citable format. It would also be a way for instrument techs and lab managers, who might not be publishing regularly, to get some additional credit via citable works. For researchers with NASA funding, MoonDB may suffice to meet the newly implemented Data Management Plan requirements.

The topic of unpublished datasets drew the most discussion at a MoonDB lunch/seminar during the Lunar and Planetary Science Conference in March 2016. Quality control was probably the biggest concern, especially when discussing the possibility of future versions of MoonDB incorporating geochronology datasets. Without context, you can calculate just about any date you want from a sample set (but really, this applies to published values as well). To combat this and other potential issues, unpublished datasets will be subject to internal review, and/or may be sent to external reviewers. In some cases, these may end up being more closely scrutinized than if they had been published online as supplementary material to a paper. On the subject of context, there is a potentially powerful (if it gets used) ability to link analyses with specific locations to sample photos (e.g., on thin sections). 

MoonDB is still seeking feedback from lunar researchers on community needs. Those who have previously submitted data to PetDB or earthchem could be especially helpful in spotting customizations necessary for lunar datasets.

Visit MoonDB.org

Advertisements