Environmental Diagnostics at Fort Union National Monument, Part 2

In June, the Penn Architectural Conservation Lab, under the direction of Frank Matero, assembled a team to revisit Fort Union National Monument. Each member of the team, comprised of current graduate students as well post-graduate fellows (myself included), had a specific focus. Topics ranged from surveying and analyzing the current structural bracing present on the site and the pathologies involved with the structural behavior of the adobe walls, to recreating historic photographs to identify possible coursing patterns–an allusive detail on many of Fort Union’s walls, and such a critical structural insight. All topics were motivated by the same focus on climate change and the impacts on climate-sensitive historical structures, such as adobe construction in the southwest.

The initial task for all members was to join forces and carry out a rapid assessment of the walls and wall segments of the mechanic's corral, which is one of the more prominent architectural highlights of the compound. The form had been prepared prior to the visit in Microsoft Access.
The initial task for all members was to join forces and carry out a rapid assessment of the walls and wall segments of the mechanic’s corral, which is one of the more prominent architectural highlights of the compound. The form had been prepared prior to the visit in Microsoft Access.

My particular focus was to validate, as best as possible, the preliminary findings from my previous report on environmental diagnostics & simulations of the mechanics corral ruin and the to identify and prioritize walls for monitoring and further focus. IR thermography was used on the walls, both at sunset and sunrise, to try and identify possible temperature differential resulting from moisture retained in the walls. What was discovered, actually, was that much of the shelter coat applied on the walls throughout is detached, creating unique blotches on the IR signatures where the shelter has possible detached. Both IR photos and video were taken. Thermodynamically, when the shelter coats detach, they have such little cross-section compared to the wall mass itself that they heat up, cool down, saturate, and dry dramatically faster–resulting in exacerbated deterioration–which, the remnants of, it was found, cause a number of resulting damaging mechanisms (soil/debris deposition at the base of the wall, leading to moisture bridging over the foundation stones, for example). And prior to detaching completely, the space behind the detached layer is hypothesized now to restrict the beneficial effects that the shelter might have toward aiding in drying the adobes, and actually trap moisture vapor instead. Further hygrothermal analysis will be done prior to our next visit to confirm this theory.

Certain portions of walls pre-selected for monitoring, based on earlier research done in the ’90s by Oliver & Hartzler, were exposed (meaning the shelter coats were carefully removed, to expose the underlying adobe) to both investigate the bond patterns of the masonry and to capture IR photos of adobe just as the coat was removed–possibly revealing trapped moisture. No indications of this kind were found, except for when the exposure was taken in an interior corner, which geometrically will retain more moisture than an exterior wall. Also, to try to gain a better understanding of the moisture-retaining properties of the walls, we performed in-situ moisture tests with a ‘speedy moisture meter,’ a system that measures the moisture of a material by measuring the reaction it as with calcium carbide. This reaction takes place within a sealed container, in which the moisture reacts with calcium carbide and produces acetylene, the built-up pressure of which can measured and correlated to moisture content by unit weight.

As surveying the wall caps was a particularly important focus of our time onsite, it was imperative to formulate a better methodology for processing the photos (taken by two teams with 25′ painters poles and attached GoPros). While photos are beneficial in their own way, we decided to experiment with photogrammetric methods onsite using Agisoft PhotoScan.From this model it was possible to export cross sections and wall profiles, which when validated in the field were surprisingly accurate.

The ACL is planning to return to the site in early November to carry out further analysis and to install a monitoring methodology, which is currently being developed. More to come on that. Stay tuned!
20160627_171345_hdr


Leave a Comment