Growing Over a Heart in Ruin
The Digital Tools that will Restore Notre Dame
I wanted to see you again, touch you, know who you were, see if I would find you identical with the ideal image of you which had remained with me and perhaps shatter my dream with the aid of reality.
― Victor Hugo, The Hunchback of Notre Dame
In the 850 years the Cathedral of Notre Dame has sat on the banks of the Seine it has undergone numerous desecrations, amendments and repairs; its iconic flying buttresses were invented and added one-hundred years after its initial construction, its sculptures beheaded in the French Revolution while the building was repurposed as a food warehouse and the most recent iteration of its spire a revision undertaken in the 1830s, when on the success of Victor Hugo’s The Hunchback of Notre Dame popular opinion pushed King Louis Philippe to undertake a massive restoration of the dilapidated building. When Eugène Viollet-le-Duc and his associates began their restoration in 1831, they had no more than archival drawings and engravings to pursue their work with. Using a large team of artisans over a span of a quarter century they restored the Cathedral as best they could, taking creative liberties when certain elements remained unclear.
As the ashes still smolder from the massive fire that consumed much of the cathedral’s roof and the Viollet-le-Duc designed spire, the restorationists tasked with rebuilding this profound masterpiece will have much more precise information to go on than their predecessors. The first step will be examining documentation of the existing conditions prior to the fire. This information is not held in hand drawings but in a digital model made by the late Dr. Andrew Tallon in 2015. Tallon scanned the entire building using LIDAR (Light Detection and Ranging) to create a Point Cloud, a 3D model constructed of billions of points that make up a model so dense that one can use them to measure conditions to the ¼ an inch. Using this digital archive of the building, preservationists will be able to see precisely where things were before the fire, down to every crack in the rib vaults. Indeed Dr. Tallon’s work documenting the building helped expose numerous flaws in the construction, flaws that ironically may have been part of the planned renovations being undertaken at the time of the fire.
What Dr. Tallon’s model won’t do is coordinate the rebuilding effort. The scans, however precise, don’t hold much more than location data. They are to the construction process what a taxidermied animal is to a zoologist; a specimen severely lacking in depth or context. To rebuild a piece of Architecture as massive and meticulous as Notre-Dame de Paris, the team will need more than just a 3D model; they need controlled views such as plans and elevations that are clearly documented to use in the field during construction, they need a precise definition of where one building material starts and another ends — for instance Viollet-le-Duc’s now destroyed spire was a detailed 750-ton construction of oak covered with lead — and they need to translate the amorphous blob of billions of points into better defined building objects such as doors, windows and bricks.
To do this one needs a Building Information Model (BIM), which is a next generation tool of Computer Automated Design (CAD). Where Architects once simply used computers to draw a straighter straight line, they are now using software in a far more advanced way, creating models that are also databases, that document not only materials but also volumes of interior area, phases of construction and the inevitable revisions that arise during the messy process of building. While it is unclear if a BIM exists of Notre Dame, the practice of creating one of these 3D databases has been undertaken on several historic buildings in Rome. When an existing historic building leverages Building Information Modeling it is called HBIM or Historic BIM. Still in its nascent stages, HBIM has been used in the US for properties such as George Washington’s Estate at Mount Vernon, providing an opportunity to keep a digital reference of the building both for those who can’t make the pilgrimage and as a digital backup for future restoration.
No one can anticipate a disaster on the level of the Notre Dame fire. It is rare for such a well-protected building to undergo such a destructive event. Still there is value in a HBIM model for more mundane work as well. The model can double as a digital twin for the building, giving the facility managers a tool for planned restorations as well as documenting completed restorations. It becomes the record of the building in a way an Xcel spreadsheet never could. Architecture, unlike the Tech Industry, moves at a glacial pace. If that weren’t enough, budgets of historic buildings are often tight and used primarily for physical restoration tasks, such as cleaning a façade or repairing a structurally unstable column. It is not unusual for the main source of documentation on a historic building to have more in common with the documents used in the 1831 Notre Dame restoration than the ones that will be used for it’s 21st century rebuild. The hope is as technologies evolve and costs decrease, Laser Scanning will come to be seen as a more accurate and cheaper alternative to someone measuring and producing drawings by hand. Wider adoption will likely result in more accurate and efficient restoration efforts. For now however, it is the rare exception such as a Notre Dame that has been thoroughly documented digitally.
The Cathedral of Notre Dame will certainly be rebuilt and to a far more accurate representation than the previous restorations in its nearly millennium long history. Unlike in 1831, these 21st century restorationists will not need to take liberties in the reconstruction –they will have a perfect replica from which to reference, made of a billion digital dots.
Dan Edleson is the Principal of STEREO, a Building Information Modeling firm focused on digitizing existing building conditions. www.stereobim.com
