Home FAQ PEMdata Demo Monitoring Support Purchase Contact

Monitoring

It is important to understand how the last several decades of preservation research (at IPI itself, and in many other institutions) have fundamentally altered basic approaches to analysis of temperature and humidity conditions for collections. In the past, the paradigms for environment control were based on target ranges of temperature and RH, and analysis consisted mostly of determining whether specific conditions were being met. Deviations from 'ideal' ranges were by definition undesirable. 'Flat lining' (absolutely unchanging T and RH) was seen to be the best for all collections, and had the added advantage that it was easy to look at a graph and tell if the lines on the graph were flat or not.

Research that closely studied how materials deteriorate showed that a single environmental condition could never really be ideal—not for the different components of a composite object, let alone for a collection consisting of many different kinds of objects.

Every environment is a compromise in which various threats operate to greater or lesser degrees depending on the nature of the object and the prevailing conditions.

Ideal conditions for minimizing metal corrosion should be quite dry, while dryness shrinks wood and leather, threatening cracks and tears. Organic materials decompose at a rate that depends heavily on temperature, so the cooler the better for them. Some moisture is necessary to make parchment flexible, though too much moisture grows mold and causes permanent deformation.

Environmental threats are a continuum, not an on/off phenomenon. The simple notion of one-size-fits-all environments does not fit the scientific facts of collection deterioration. A more nuanced view, in which individual mechanisms of deterioration are considered separately, leads to much more effective environmental management.

The air inside buildings is essentially outdoor air. It is cool and contains very little moisture in winter, while in summer it is warm and contains lots of moisture. In real life, HVAC systems smooth out some of the influence of outdoor conditions, but almost never do they produce the kind of flat lines that many people seem to expect. Novice data analysts almost always are surprised by how much variation appears on temperature and RH graphs. The key question in analysis of data is not whether fluctuations exist (they nearly always do) but what do they mean? Daily and weekly fluctuations are usually smaller in magnitude than seasonal ones.

The second big problem with targets and flat lining is that, with a few rare exceptions, no one can achieve flat line conditions.

Digging a little deeper into the historical background for ideal targets and flat lining reveals how these ideas came to be regarded as orthodoxy. Beginning around WWII, major art museums became more aware of the risks of excessive dryness and dampness with paintings and furniture. The conservation profession began to emerge and practitioners wrote about the necessity of controlling RH to avoid acute physical problems due to improper RH. They advocated monitoring of T and RH conditions and recommended an RH range (near 50%) that minimized the risks of dryness and dampness to fine art collections. Temperature was less important for these kinds of physical risks to fine art, though temperature is more significant with library and archive collections. Gradually, consensus recommendations for 'ideal' conditions came to be 20°C (68°F) and 50% RH, with little permissible variation.

Seasonal changes are the most significant in determining the preservation quality of collection environments. Winter dryness and summer heat and high humidity are the most difficult preservation environment challenges.

Although even the early conservators themselves recognized the gross oversimplification that articulating an 'ideal' like this represented, a number of circumstances helped to shape the popular acceptance of these concepts. Monitoring of conditions (if done at all) was with weekly pen-and-ink charts made by hygrothermographs and without the power of computers to do extensive calculations or condense a whole year of data onto one graph. As a result, seasonal variations were hard to notice. Monitoring with weekly charts enforced a short-term perspective that worked against perceiving longer-term, more significant trends. Without computer analysis, no derived statistics or metrics were possible without extremely laborious hand calculations. Whether lines were flat or targets were achieved was about all one could learn from monitoring in this way.

Because the focus of concern was the effects of RH on fine art materials, the importance of temperature was diminished, and it did not seem to matter too much if the 'ideal' conditions were crafted to be at temperatures comfortable for humans. (Later research shows that cool temperatures are essential for some materials and beneficial to many, including the organic components of fine art objects.) Gradually, steady RH near 50% at temperatures comfortable to humans came to be regarded as the very best conditions possible and the self-evident basis for assessing environmental quality. Everyone likes simplicity, and everyone likes to be comfortable. But it is important not to ignore the real influences of environment on collections and to confuse human comfort with the needs of inanimate collection objects.

IPI's research has focused on developing ways to quantitatively estimate the effect of environmental conditions on specific mechanisms (pathways) of deterioration. Based on its research, IPI has proposed a set of environmental metrics that can be used to assess and manage collection storage environments.

If striving for flat lines on graphs and controlling to a single 'ideal' condition is not very practical or scientifically justifiable, what is the best approach to data analysis?

The metrics are quantitative numerical estimates of the rate of environmentally-induced decay in collections, broken down into specific numbers for the risk of mold growth, physical damage, natural aging in organic materials, and metal corrosion. These estimates are derived from the observed T and RH data using algorithms created by IPI. Each metric integrates spans of time into a single value representing how the environment is likely to affect one particular form of deterioration, taking into account all the ups and downs of temperature and RH.

With targets, you are in or you are out. If you are out, there is little guidance on what that means, or what to do about it. It is particularly difficult to 'eyeball' the shape of T and RH lines on graphs and weigh the significance of deviations, attempting to balance their magnitude (how far from ideal) with their duration. Add to this the facts that both temperature and RH play a role, and that relative importance of temperature vs. humidity is different for various forms of collection decay, and the 'mental arithmetic' needed for eyeball analysis becomes very difficult.

This ability to integrate all the fluctuations in temperature and humidity into a single overall estimate of decay rate is a powerful feature for analyzing data.

The IPI Metrics consist of TWPI for natural aging, MRF for mold, and DC/EMC. One of their best features is that they always analyze data in the same standardized way. When there are many spaces to manage, it is extremely helpful to have quick, automated analysis. The metrics are very useful in flagging potential problems from a quick screening of the data. Although the metrics may be unfamiliar at first, after a little time working with them, their value becomes apparent, especially when there are a number of datasets to analyze.