Rescuing 50 Years of Apollo Sample Data
NASA’s Johnson Space Center stores its moon rocks in a big vault behind a water-tight door. To go inside, Kerstin Lehnert and her team had to cover themselves from head to toe with cleanroom clothing, then walk through pressurized air to blow off any remaining contaminants. Inside, the lunar samples are sealed in Teflon packaging and protected in locked metal cases.
Lehnert, a geoinformatics specialist at Columbia University’s Lamont-Doherty Earth Observatory, was impressed with the facilities. But for her, they also underlined a huge discrepancy: a lot of the data that has come from studying these lunar rocks is at risk of being lost or has been lost already. Much of it has never been published; instead, it languishes, forgotten, in laboratories and offices around the world, or gets discarded when older scientists retire. The data that is shared publicly is spread out in hundreds of papers and abstracts, often in un-usable forms.
“It’s really mind boggling that the results of these studies have not been as well-preserved as the samples,” said Lehnert. After all, “the samples are curated in order to generate the data.”
But Lehnert and her team have started correcting those past mistakes. She and her colleagues have been working on MoonDB, a database that has restored and rescued 90 percent of the analytical data from the Apollo samples. Now, with additional funding from NASA, they will develop the Astromaterials Data System, which will become the home of the MoonDB database. Over the next five years, the Astromaterials Data System will collect, organize, and digitize study results not only from the samples collected by the Apollo missions, but also studies based on meteorites, cosmic dust samples, and samples from asteroids and comets gathered from outer space.
NASA’s previous attempts to compile Apollo mission data resulted in a large number of PDFs with limited usefulness, since researchers couldn’t sort or interact with the data.
In contrast, Lehnert aims to build a database that makes it easy to access, search through, and visualize the information, enabling data mining and statistical analyses. She and her colleagues have gone through the literature to identify all of the Apollo studies from the last 50 years. Curators on the team track down the original data, then digitize it and gather other information such as the data quality and how it was collected and calibrated — essential info for other researchers who might wish to use the data. The team has already compiled data from about 800 publications, and in the coming years they hope to build the database.
“By putting all of these data into a single database, we’re putting the puzzle pieces together into a big picture — you can see things that you couldn’t see before.”
When it’s finished, the database “will allow completely new views and comparisons of the data,” said Lehnert. She compared it to the PetDB database that was developed at Lamont-Doherty in the 1990s and revolutionized access to data about the composition of terrestrial rocks by organizing and synthesizing it. “Lots of people study a particular rock type or a particular region, and by doing this they create a piece of a puzzle,” Lehnert explained. “By putting all of these data into a single database, we’re putting the puzzle pieces together into a big picture — you can see things that you couldn’t see before.” The AstroMat database is expected to have similar benefits for the lunar and planetary science communities.
Earth Institute director Alex Halliday, who studies lunar samples to determine where, when and how the Moon formed, said the database will save researchers a lot of time and effort, and could help to reconstruct the earliest histories of both the Moon and Earth.
Currently, he said, “it is hard to piece together all of the information for each sample without trawling through publications and on-line summaries…. [The database] means I can search efficiently and effectively for the best sample to tackle a particular problem I want to study. It means I can more readily test theories for how the Moon formed and differentiated by easily pulling together the critical data needed to demonstrate consistency between theory and observation. It means I can efficiently compare my data with those obtained by others.”
The database will also help to protect what’s left of these precious materials from outer space. “NASA is worried about overusing these samples,” said Lehnert. “There are limited quantities available. If they don’t have a good record of what studies and projects have already been done, they might be handing out material unnecessarily.”
Making this invaluable information more open and accessible will also help to ensure that it benefits of all of humankind, and gets preserved for the next generation of scientists and explorers.
Ultimately, the database could even help in informing future missions to the Moon and locating new rocks to bring back to Earth, Halliday pointed out. “As we plan return missions to the Moon, we can use the database to help us identify good landing sites for tackling important science questions.”