From the Torch to the Toes, Digital Insurance
Published in The New York Times
September 11, 2003
How the Statue of Liberty could be recreated, after a disaster
By FRED A. BERNSTEIN
ONE HUNDRED AND TWENTY years ago, the Statue of Liberty sat in packing crates; France had created it, but there was no money to erect it. Then Joseph Pulitzer used his newspapers to shame Americans into contributing the needed $100,000.
Elizabeth Louden knows how Pulitzer felt: her 21st-century construction project -- creating a ''digital map'' of the Statue of Liberty -- is out of funds. During her last visit to the statue, in August 2001, Ms. Louden and her team from Texas Tech University in Lubbock used a laser scanner to record the location of 94 million points on the statue's surface. But they had no way to scan the top of the head and shoulders and other parts not visible from the ground.
So she is busy writing grant proposals. Ms. Louden, an associate professor at Texas Tech's architecture school, says she can complete the project for $100,000.
It is an investment that she hopes will never pay off. The detailed drawings that she has promised to produce would probably be used to recreate the statue in the event of a terrorist attack.
Although the project was initiated in early 2001, before the Sept. 11 attacks, Paul Dolinsky, the National Park Service official who commissioned it, said he had from the outset envisioned such drawings as ''an insurance policy on the building.''
''These days,'' said Mr. Dolinsky, chief of the service's Historic American Buildings Survey, ''in the back of everyone's mind, whether they want to admit it or not, is the thought of catastrophic replacement.''
Mr. Dolinsky's agency, known as HABS, was founded 70 years ago. But in the last two years, he said, ''there's been a lot more awareness throughout the government of the significance of what we do.'' Still, he contends, with a budget of only $2 million a year, ''we're practically a mendicant order.''
So HABS relies on professors and graduate students to do most of its field work. Mr. Dolinsky chose Texas Tech for the Statue of Liberty project largely because John P. White, who teaches courses in historic preservation there, has been working with the buildings agency for nearly 30 years, and the school has a $200,000 scanner that the agency cannot afford.
That scanner, made by Cyra Technologies of San Ramon, Calif., is about the size of a toaster oven. The device, called the Cyrax 2500, directs a laser at a target, recording how long it takes for the light to return. But, unlike a simple digital surveying instrument, the Cyrax ''pulses'' its beam in 800 bursts per second, releasing each burst at a slightly different angle. Software on a laptop attached to the Cyrax uses the data to produce a three-dimensional map of the target surface.
Ms. Louden said the machine was designed for industrial applications, like mapping a refinery so that an oil company could figure out where to install new pipes. In such a situation, knowing how a building was built is more important than knowing how it was designed, she said.
The same is true in the world of historic preservation. To restore or recreate a building, ''you need drawings of what's there, rather than what, theoretically, is supposed to be there,'' Mr. Dolinsky said.
In the case of the Statue of Liberty, the flowing copper skin was not built from blueprints, but by taking measurements from a scale model with calipers. As a result, ''there were no existing exterior drawings,'' he said. (Drawings of the interior were created during a renovation of the statue in the 1980's.)
Mr. Dolinsky considered other ways of creating the exterior drawings. One would have been to take physical measurements, a long and possibly dangerous task involving ropes and scaffolds.
Another idea was to extract measurements from digital photos. But Mr. Dolinsky said, ''A photo won't give you the level of detail you would need to reconstruct a piece of the statue's clothing.''
Then in 2001, Ms. Louden, Mr. White and their colleague Glenn E. Hill agreed to provide HABS with a complete set of elevations and cross sections of the statue by using the scanner. Visiting Liberty Island in August 2001, the team took measurements from 13 points around the statue's base. The Cyrax was set to measure the statue at least once every quarter-inch.
But in their four days on the island, the team was able to scan only 70 percent of the statue's surface, Ms. Louden said. Some parts of the statue were invisible from the ground. Other sections, notably the torch, swayed in the wind, defeating the scanner.
A month later, after the terrorist attacks, the Statue of Liberty was closed to the public. (It remains closed, although the grounds of Liberty Island were reopened in December 2001.) Creating a digital Statue of Liberty took on a new urgency.
But two years later, the three professors are still manipulating data. They attribute the delays to their teaching schedules, a lack of money to pay graduate students, and software limitations.
The first task after returning from New York was to merge the 13 sets of raw three-dimensional digital data, known as point clouds, by identifying where they overlapped. The computer helped, but ''we had to point it in the right direction,'' Mr. Hill said.
The next step, he said, was to proceed from points to surfaces, literally connecting the dots to turn each cloud into millions of triangles in a process known as tessellation. Then software called Geomagic was used to ''draw'' a smooth surface over the triangles.
But tessellated surfaces are a far cry from the elevations and sections that architects depend on. The team had selected a computer-assisted design program to create the drawings, but importing the data created a bottleneck. ''We were are trying to stream 94 megabytes, and the computer kept freezing,'' she said. ''We spent months trying to figure out how to avoid crashing the computer.''
Eventually, Mr. Hill said, the team created smaller, more manageable files corresponding to sections of the statue.
That was last spring. But the graduate students doing most of the work need to be paid, and the HABS grant has been spent. ''We finally have a plan that works,'' Mr. Hill said, ''and we're out of money.''
Another challenge is to map the 30 percent of the statue's surface that Ms. Louden missed on her last trip to New York. Capturing the statue's feet will require placing the Cyrax on a platform. Meanwhile, she is working on obtaining data on the statue's pate from digital photographs.
With the adoption of new technologies, Mr. Hill said, plans and elevations might not be needed to reconstruct the statue. Computer-controlled three-dimensional ''printers'' could potentially generate sections of the statue directly from the point clouds. But the team still plans to give HABS the documents.
Ms. Louden explained: ''HABS is looking for materials with a 500-year shelf life. Even though we hope that the digital file will be readable far into the future, we are relying on a traditional method of communication, which is the hard copy.'' Typically HABS drawings are available to the public through the Library of Congress.
''We'll finish during this academic year,'' Ms. Louden said. Even if no money comes in, she said, ''we'll find a way.''