— In a brightly lit room on the third floor of the Museum of Natural History here, stacks of wooden drawers are covered in glass, some panes so dusty that it is difficult to discern exactly what’s inside. When the glass is removed, rows of carefully pinned insects are revealed, gleaming in brilliant colors like precious jewels.
The biologist Alexander Kroupa plucks an amethyst-colored beetle from the drawers with metal pincers. “Amazing, right?” he said. “As beautiful as the day they were collected.”
Mr. Kroupa and 14 colleagues are in the midst of a vast undertaking: digitizing and publishing online the museum’s entire collection of insects, including high-definition three-dimensional images of thousands of particularly important specimens. The researchers here are not alone. Museums around the globe are trying to harness the power of digital technology to make available collections that have long lay dormant on shelves and in dusty cabinets.
For years, scientific institutions have scanned images of specimen drawers or individuals in their collections. But technical advances provide new opportunities to create extraordinarily detailed images and data that may be critical to answering some of the biggest questions in conservation biology, experts say.
By looking at the historical distribution of a species as revealed in the collections, for example, scientists can determine how the ecology of a region has changed because of industrialization, human settlement or climate change. The old collections provide deep insight into changes in genetic diversity, crucial for saving a species, informing urban planning or protecting vital food resources in the future.
“To really understand biodiversity, you have to look in the museums,” said Brent Mishler, a professor of plant biology at the University of California, Berkeley
The digitization efforts also are helping scientists create a permanent record of the natural world in a period of rapid extinctions.
“This is one way of documenting what we are about to lose,” said Quentin Wheeler, an entomologist and president of College of Environmental Science and Forestry at the State University of New York
. Digitizing museum specimens will reveal “irreplaceable clues needed to reconstruct evolutionary history, to understand where we and all the other species came from.”
In Berlin, the natural history museum’s collection consists of more than 35,000 drawers holding about 50 million individual specimens. So far, the team has scanned about 10,000 drawers. Some specimens are selected for high-definition scanning, which will allow scientists from anywhere in the world to examine these colorful creatures in remarkable detail, even to maneuver them on screen for close scrutiny.
“We want everyone to see them, the public and researchers, to see what’s in the collection,” said Bernhard Schurian, an imaging specialist at the museum who is overseeing the technical aspects of the project.
It is slow, painstaking work.
Each specimen drawer must be carefully removed from one of hundreds of alder wood cabinets filling six floors of the museum. The drawers are dusted off, and a small brush sometimes used to tidy each delicate specimen.
Some of them are type specimens — individuals used to characterize their species. An animal is judged to be a member of a given species by its resemblance to the type specimen, the gold standard.
For centuries, scientists who wanted to study a particular type specimen had to visit the museum where it is kept or have the specimen sent to them. Either way, the potential for damage was high: fragile body parts would sometimes fall off during inspection or transport, causing irreparable damage.
Each type specimen is “like the Mona Lisa,” said Katja Seltmann, a biologist at the American Museum of Natural History in New York who specializes in biodiversity informatics. “If an antenna or a leg breaks, all of a sudden, a really large part of information about that organism is gone.”
Other efforts to digitize type specimens are underway. The Atlas of Living Australia
harnesses the contributions of individuals to catalog every known species on the continent. The Encyclopedia of Life, brainchild of renowned biologist E.O. Wilson, has published more than 1.2 million pages with 3.2 million images of species.
Generally, these projects involve — at most — making top, bottom and lateral images of specimens. But the Berlin museum is taking the idea to a new level.
In a cramped, darkened room in the basement, individual type specimens are placed on a rotating drum in a lightbox and photographed at many angles with a macro lens. The team uses computer software to stitch the photographs together into a single focused image, which can be downloaded and viewed at up to a 100 angles.
Depending on the size of the insect, as many as 500 images are taken at a single angle; 3,000 to 5,000 images are taken of a single specimen. The resulting data can add up to more than 100 gigabytes, far too large to download from the Web.
The team relies on compression and an algorithm similar to that used by Google Maps to load only a few portions of the image at a time on an individual screen. The result, called ZooSphere, is a magnificently detailed picture of the type specimen, available to anyone anywhere.
The researchers here are focused on insects, but other institutions have been trying to digitize vast collections of plants, mammals and fish. Few have attempted to provide this level of visual detail.
One of the largest efforts in the United States
, called the Integrated Digitized Biocollections, or iDigBio, is currently underway at the University of Florida
. Funded by the National Science Foundation, iDigBio has brought together 283 institutions in all 50 states, mostly museums and universities, to create a common language and process for digitizing images of species.
“Everybody knows there’s a tremendous amount of information in natural history collections,” said Larry Page, the curator of fishes at the Florida Museum of Natural History and the director of iDigBio. “But the collections are inaccessible to virtually everyone. Even scientists working on particular groups of organisms don’t know what’s contained in the other museums.”
Currently, the iDigBio search portal contains 45 million specimen records and about 12 million images. But that is not nearly enough, Dr. Page said. He estimates there are half a billion specimens in the United States alone.
The project, in its fifth and final year of funding, has requested support for another five years. Dr. Page says the portal will prove an invaluable resource for scientists.
In a few years, he said, a scientist studying climate change might be able to glean information from the iDigBio database about how an insect’s range has changed and make predictions about future trends.
“The data tell us what occurred where at certain periods of time, and we can link that to environmental data and then look at the changes they anticipate,” Dr. Page said.
Right now, the iDigBio project is focusing mostly on building of a vast database of species records, information collected by scientists in the field about when and where a particular species was found. Initially, the project did not place much emphasis on images, but that has changed.
“I was skeptical about use of images when the project first started,’ he said. “I thought database information was enough. But images are becoming of huge interest to everyone because there is so much information in them.”
That is why the efforts in Berlin and elsewhere have such potential. The high-definition imaging of type specimens and specimens in general opens up the natural world in a way that scientists only dreamed of a decade ago. Science stands to benefit not just from the advances in knowledge gained from worldwide access to these specimens, but also by encouraging students to pursue careers in biology or environmental work.
“This is a chance to make everyone aware of how valuable the data are in natural history collections, how much they tell us about the natural world,” Dr. Page said.