FALL 2018
Collection Environments and Evidence-based Decision-Making
By Roman Kozlowski
conservation image

IN 2014 The International Institute for Conservation of Historic and Artistic Works (IIC) and the International Council of Museums-Committee for Conservation (ICOM-CC) issued a joint declaration on environmental guidelines for museums, generally perceived as a fundamental milestone in advancing the debate on appropriate environmental specifications for collections.1 The declaration stated that the issue of collection and material environmental requirements is complex, and conservators/conservation scientists should actively seek to explain and unpack these complexities." It called for a more customized approach to setting the environment for collections and historic interiors, taking into consideration the different requirements needed for objects on display, in storage, or in transit, as well as any individual sensitivities to certain conditions and the degree to which objects may have become acclimatized to their local environment.

Moving to variable specifications requires evidence-based decision-making. This entails establishing clear communication between research and heritage managers, and an effective transition from basic research to application. A crucial research focus is climate-induced change on humidity-sensitive materials, an area the Jerzy Haber Institute in Kraków has been active in for a number of years.

Climate-induced damage to humidity-sensitive materials is an important risk in most museum collections and historical interiors, as such materials undergo physical change when they lose or gain moisture. The constraint from free movement, due to rigid construction or connection to materials that respond differently, induces stresses in the objects. These stresses can cause deformation, cracking, and delamination. Heritage science and conservation practice have developed two general approaches to providing evidence necessary to inform climate specifications: (1) analyses of the historic climates to which the objects have "acclimatized," and (2) analysis of the physical response of materials and objects to relative humidity (RH) and temperature fluctuations.

Awareness of object acclimatization to a particular indoor environment has been long reflected in the requirement by some that climate conditions be retained as fully as possible when vulnerable objects are moved from their usual location for restoration or exhibition. Stefan Michalski coined the term "proofed fluctuation," defined as the largest RH or temperature fluctuation to which the object previously has been exposed. He assumed that the risk of further physical damage from fluctuations smaller than the proofed values is low if the object and environment are not altered. If the past fluctuation was enough to cause fracture, the object has fractured, and the cracks reduce the stress that would otherwise develop in the undamaged material. The acclimatization concept has been convincingly confirmed by two new research tactics. The first is precise direct tracing of damage in objects using acoustic emission (AE). The second is collecting observations from a large group of well-defined objects (exemplified by the Rijksmuseum study on the effects of humidity fluctuations on decorated wooden panels in the museum's collection).

The AE method—based on monitoring the energy released as sound waves during fracture processes in materials—was successfully used in over a year of monitoring crack propagation in wooden elements of two pieces of furniture, in the National Museum in Kraków and the Victoria and Albert Museum in London, selected by conservators as particularly vulnerable to damage potentially induced even by the controlled environmental conditions in the galleries.2 The crack propagation determined was below 1 mm per year in each piece of furniture—a minute change for any practical assessment of damage, which could be recorded only owing to the amazing sensitivity of the AE sensors.

conservation image

In the Rijksmuseum study, construction, material properties, and condition of more than 370 decorated oak panels (cabinet doors and panel paintings that had been in the same location for about 100 years) were examined.3 Shrinkage cracks and failing joints were the common types of damage and were generally formed before the objects entered the museum collection. The uniform damage pattern reflected early "acclimatization" of similar wooden constructions to probable large RH variations in the uncontrolled environments in which the objects were historically kept.

The acclimatization concept was explicitly expressed in standards on control of the indoor environmental conditions. Among them, the European Standard 15757:2010—Conservation of Cultural Property–Specifications for Temperature and Relative Humidity to Limit Climate-Induced Mechanical Damage in Organic Hygroscopic Materials—is widely referred to by museums and research institutions. For example, out of nineteen presentations at the Climate for Collections: Standards and Uncertainties conference organized by the Doerner Institute in Munich in November 2012, eleven quoted the standard as a reference. The standard provides a methodology of processing accumulated past climate records to establish more quantitatively targeted microclimates, specifying average levels of climatic parameters and their seasonal drift, as well as bands of tolerable short-term fluctuations. The standard stresses that the harmlessness of the existing climatic conditions has been a key assumption in the acclimatization approach, which has to be carefully checked in each case.

Obviously, environmental specifications cannot be based on the acclimatization approach when:
1. new damage continues to accumulate in an object;
2. an object has to be moved to a different climatic environment; or
3. conservation treatments may alter the safety margins of objects
achieved by their acclimatization to the past conditions.

In these instances, decision-making requires analysis of moisture and mechanical response of materials and their assemblies to address how much variation in RH is actually safe for a specific object made of hygroscopic materials.

A particularly effective way to analyze the response of materials is computer modeling, which simulates "real-time" moisture movement and the resulting strain and stress fields across objects of varying shape, thickness, or water vapor permeability in response to RH variations. The modeling has been based on existing and developing information on material properties such as adsorption or desorption of water vapor, moisture-related swelling and shrinkage, water vapor diffusion and surface emission coefficients, and tensile properties. The modeling provides a quantitative assessment of the climate-induced risk based on the analysis of moisture-related dimensional response of objects. By offering a direct rather than indirect measure of the hazard, significant progress in the practice of evaluating climate parameters by themselves can be made.

A new online environmental data analysis tool, HERIe, is being developed collaboratively by several institutions including the Getty Conservation Institute (GCI), to overcome a barrier to exploiting advantages of modeling of object moisture response in conservation practice.4 The user provides basic characterization of objects in the specific collection and uploads RH data, recorded in a gallery or simulated for various climate-control scenarios. The software processes the data, using the precalculated database, into strain versus time history experienced by the object. The risk of damage is then assessed by comparing strain against a critical level selected by the user—the damage criterion.

Recent years have brought promising new initiatives that address the major deficiency in the modeling—the lack of material properties and failure criteria (e.g., the point at which movement in wood or paint layer exceeds their elastic limits). These are derived directly from investigations and observations of actual objects aged and adapted over decades or centuries to indoor environments in which they have been preserved. Such adaptation might have involved an unknown level of permanent change, like deformation or fracturing, making historical materials different from new materials, also with respect to their vulnerability to damage processes.

conservation image

The new initiatives range from mechanical characterization of aged materials by monitoring deformation of historical painted panels (also in laboratory tests developed by GESAAF [Dipartimento di Gestione dei Sistemi Agrari, Alimentari e Forestali] of the University of Florence), to using micro- and nano-indentation methods on submillimeter samples of aged paint layers, glues, or gessoes, developed by the GCI within its Managing Collection Environments Initiative to better estimate and, ultimately, improve accuracy of numerical modeling predictions. The micro-indentation technique, which permits multiple measurements of a single sample, will be used by the GCI to determine moisture-related mechanical properties of historical materials contained in glue paint decorations of known origins and detailed aging histories in Norwegian stave churches. This research is part of the newly initiated SyMBoL project (Sustainable Management of Heritage Buildings in a Long-Term Perspective), coordinated by the Norwegian University of Science and Technology in Trondheim.

Another equally important area for future conservation efforts is refining damage criteria to base them on the observation and monitoring of actual objects displayed in real-world conditions in museums and historic buildings. Again, no matter how much evidence is obtained from laboratory work with replicas simulating original objects, decision makers will remain skeptical of the evidence obtained. Well-documented damage development on freshly treated or consolidated objects would be of interest, as such objects may be particularly vulnerable to RH variations.

Finally, an acute gap in the conservation literature is the general absence of thoroughly documented reports on the effect of climate control failures on actual objects. These reports could cover collection observations in which damage to objects could be related to unusual recorded humidity variations resulting from power failures or poor maintenance of air-conditioning equipment, insufficient response of the systems to sudden spells of extreme weather outdoors, or incidents of water intrusion into display or storage areas. Hannah Singer's detailed report on the effect of dramatically increased RH levels on the paper collection in Vienna's Albertina Museum due to rainwater infiltration remains, unfortunately, an exception.5

The emerging methodology linking collection analysis and numerical and experimental studies to better understand climate-induced damage of specific objects has so far focused on objects of fine and decorative art—and, predominantly, decorated wood. An obvious next step will be evidence-based climate specifications for library and archival collections that contain almost exclusively hygroscopic materials: paper, board, parchment, leather, and wood. The model of moisture and mechanical response would need to address broader patterns of damage, including not only physical failure such as cracks or tears but also three-dimensional deformation like curls or cockles. The gap in solid information on the issues is evident in the most up-to-date ISO standard 11799:2015—Information and Documentation - Document Storage Requirements for Archive and Library Materials—which does not specify any recommended range of RH variations.

Coupling chemical degradation of modern artistic materials like plastics to risk of mechanical damage is another emerging field of research.

The IIC/ICOM-CC guidelines called upon conservators and conservation scientists to take a more active role in developing evidence-based environmental specifications. The latest developments offer diverse approaches in which observations and data increasingly gathered from measurements of historic materials can supply input into algorithms modeling risk of climate-induced damage and provide a frame of reference for conservation and museum professionals. We are also witnessing a decrease in the obstacles to research feeding back into conservation practice: fundamental science has become more accessible through evidence-based standards and software tools, advanced techniques of object monitoring now are more widely used for routine work, and research projects lead more often to general outcomes. However, to build on these trends, developments should be supported that aim to bring together in-depth, object-based information and experimental and modeling studies, and that enable full engagement of all actors interested in effective management of collection environments to reduce energy use while maintaining high standards of collection care.

Roman Kozlowski is head of the Cultural Heritage Research Group at the Jerzy Haber Institute in Kraków, Poland.

1. Julian Bickersteth, "IIC and ICOM-CC 2014 Declaration on Environmental Guidelines," Studies in Conservation 61, sup. 1 (2016): 12–17.
2. Michał Łukomski, Marcin Strojecki, Boris Pretzel, Nigel Blades, Vincent L. Beltran, and Ashley Freeman, "Acoustic Emission Monitoring of Micro-Damage in Wooden Art Objects to Assess Climate Management Strategies,"Insight 59, no. 5 (2017): 256–64.
3. Stina Ekelund, Rianne Luimes, Cecilia Gauvin, Paul van Duin, Andre Jorrison, Bart Ankersmit, Akke Suiker, Roger Groves, and H. L. Schellen, "The Development of a Methodology to Understand Climate Induced Damage in Decorated Oak Wood Panels," in ICOM-CC Preprints of the 18th Triennial Conference, Copenhagen, 4–7 September 2017, edited by Janet Bridgland, art. 1507 (Paris: ICOM-CC, 2018).
4. HERIe software is available at with information on the methodology used and tutorial climates.
5. Hannah Singer, "Evacuating the High Bay Racking System at the Albertina Museum in Vienna after a Water Entry in June 2009," Restaurator 31, nos. 3–4 (2010): 265–85.