For decades, climate scientists have attempted to warn society about potential significant threats related to a changing climate as an attempt to call for action. Unfortunately, many did not take the danger’s immediacy seriously, and the risks have now moved from potential to present. According to the 2021 Intergovernmental Panel on Climate Change (IPCC) Sixth Assessment Report, widespread and rapid changes in the atmosphere, ocean, and biosphere have already occurred. The predominant evidence of these changes can be noticed in extreme behavior of heatwaves, heavy precipitation, droughts, and tropical cyclones, which will all continue to increase in severity and frequency into the future (IPCC, 2023). The nine U.S. regions considered by the Fourth National Climate Assessment all have increased threats from these events into the future, regardless of which future climate scenario is considered (Black & Brown, 2020).

More than ever, cultural heritage institutions must assess their unique climate change-related threats to their collections and missions as part of their disaster and emergency management plans. According to the 2019 heritage health information (HHI) survey, 58% of institutions do not have a disaster plan, with 75% lacking both a plan and staff training to implement a plan (IMLS, 2019). Furthermore, the HHI report indicated an increased attribution of damage between 2017 and 2019 due to water or moisture (56%), with natural disasters accounting for 10% (IMLS, 2019).

The PROTECCT-GLAM project attempts to address these challenges by developing a national categorical climate change risk assessment scale for galleries, libraries, archives, and museums (GLAMs). To meet this objective, the project team must first address the lack of a comprehensive geo-referenced dataset for all U.S. GLAMs. The following paper discusses this dataset’s development and initial GIS analysis—specifically, identifying the number of GLAMs located within 5, 15, and 30 kilometers of a coastline and the number of GLAMs per capita in each state.

Previous GIS analysis of GLAMs remains limited and explores the factors at play in the location of cultural institutions, specifically public libraries, rather than the potential impacts of climate change. Wan, Chen, & Zhang (2017) sought to develop a statistical model to aid in the placement of new libraries. In their model, they account for traffic, environmental factors, a library’s constraint value, and the capacity limitation of the library. Similarly, Shorabeh et al. (2020) developed three major categories to consider when deciding where to build a new public library: centrality (population and accessibility), compatibility (air and noise pollution and comfort and security), and environmental (elevation, slope, and fault lines). Their findings indicate that optimal locations for public libraries are among areas with high population density, with ample access to transportation (nearby roadways and public transit) and security centers (police and fire stations, health resources, etc.).

Further, their results also showed that population and demographics were more critical in location decisions than accessibility, meaning that building libraries in areas with higher rates of illiterate population and student density is more beneficial than building libraries near roads or public transportation. Findings from a study investigating the relationship between accessibility and the utilization rate of public libraries support the conclusions of Shorabeh et al. (Zhao & Hong, 2023). In contrast with results from former studies, there was no evidence of a causal relationship between accessibility and utilization; in comparison, the findings showed that human factors, like motivation and customs and habits, play a more important role in public library utilization among their sample.

The location of cultural institutions has important implications for their surrounding areas and those seeking to use their resources. In particular, the existence and construction of new museums have been shown to be a catalyst for growth in the area in which they are located. For example, in a study of the proximity of museums to the city centers of Barcelona, Paris, and Turin, (Paül i Agustí, 2014) found that while most museums are located within 2 km of the city center, those outside this radius have more significant impacts on their surrounding communities than those located within the radius and can be used by community leaders to spur social and economic change within their reach. These findings illustrate cultural institutions’ role in their communities and their potential growth.

Despite these hopeful findings, however, cultural institutions often follow suit of other significant social and economic resources, meaning that areas with scarce access to these resources generally lack access to cultural institutions as well (Short et al., 2020). In their investigation of informal learning institutions (ILIs) locations across the United States, Short et al. found several ‘ILI deserts’ worthy of attention. In particular, their findings show large areas among the Great Plains, the Southeast, and the Northwest with a low relative density of ILIs compared to their population. Further, these ‘deserts’ often encompass areas with high poverty rates and low population density and disproportionately affect American Indian and Alaskan Native communities, with 0.95% of these communities lacking access to ILIs compared to the general population’s 0.10%. This problem is not exclusive to these areas, however. In a study of Washington DC, Baltimore, Chicago, and Phoenix and their respective spatial accessibility of public libraries, Cheng et al. (2021) found that, with the exception of Phoenix, socially disadvantaged populations, including non-adults, minorities, and impoverished people, have to travel further distances to access public libraries. Combined, these studies indicate that demographics facing other societal and economic hardships across the country are less likely to have easy access to cultural institutions.

To meet the PROTECCT-GLAM project’s overall objectives, the project team needed to create a dataset that included the geographic, physical location of all GLAMs in the U.S. The project team began constructing the dataset by identifying the following existing external data files:

2018 Museum Data Files from IMLS Data Catalog (IMLS, 2018)

Each external data file required different auditing and manipulation techniques to align its data to the PROTECCT-GLAM requirements. The following section summarizes the steps taken for each external data file.

After its creation for initial visualization and spatial analysis, the project team imported the dataset into ArcGIS Pro. The initial analysis focused on GLAMs most at risk for potential sea level rise by identifying all GLAMs located within 5, 15, and 30 kilometers of the U.S. coast. As noted in Table 1, a combined total of 16,877 GLAMs are within 30 km of the coast. The states with the highest number of GLAMs within 30 km are California (2,672), New York (2,566), Maryland (1,304), Florida (1,290), and New Jersey (1,159).

Secondary analysis focused on the number of GLAMs per capita in each state using the 2020 U.S. census data. For this analysis, the per capita is reported at the per 10,000 level (see Table 2).

The PROTECCT-GLAM project team successfully created a national dataset of GLAMs and conducted a preliminary analysis using ArcGIS Pro. Additional analysis will incorporate data on five variables, including hurricane wind occurrence & storm surge inundation, three feet of sea level rise, inland flooding occurrence, temperature exceeding 23-27⁰C occurrence, and dew point exceeding 9.3⁰C occurrence. The resulting data will be compiled into a five-point scale based on the average of the five variables for each GLAM.