Computing Place: The Case of City Nature
July 18, 2013, 13:30 | Short Paper, Embassy Regents E
A large proportion of digital humanities projects entail mapping and therefore computation about places. Much of that computation concerns spatial attributes and relations — location, distribution, interaction flows, and so on — the list of spatial analytic operations of interest for humanistic studies is long and growing. Much of this decidedly quantitative spatial (often spatial-temporal) analysis is performed to better understand place, a decidedly humanistic concept defined here as ‘experiential space’ or ‘space as experienced by humans.’  This distinction can be confusing because the term space is often appropriated by humanities scholars and critical theorists, in speaking for example of human-constructed spaces. In my biased view as a practitioner of geographic information science (GIScience) with a humanist bent, we need both terms and we need for them not to be conflated.
To date, computational geography (born in the late 1950s but distinguished as GIScience for a couple decades now) has concerned itself with quantitative analyses of spatial and spatial-temporal aspects of natural phenomena at geographic scales. However social scientists, and more obviously, humanist scholars, have questions concerning human experience of space and space-time. In recent years, GIScience has increasingly added methods more directly supporting qualitative analysis (Bodenhamer et al 2010; Dear et al 2011). Examples of this include (i) the semantic analysis of texts joined with spatial analysis of locations associated with their production (Cooper and Gregory 2011), and (ii) the space-time prisms of Time Geography applied to urban residents daily movements by critical human geographers (e.g. Kwan and Schwanen 2009).
In my own recent work on the City Nature project  , I have undertaken to characterize “Naturehoods” as areas within cities that are distinctive for their level of “nature-ness;” to capture the human sense of being close to nature in a classification which integrates several physical measures including satellite imagery and social variables including wealth and ethnicity. The overarching goals of City Nature have been first, accounting for the enormous variation in quantity and quality of natural areas in large U.S. cities, and then helping planners arrive at best practices for ensuring reasonable quality of life as cities world-wide grow at an unsustainable pace.
The Naturehoods profile combines satellite measures of mean distance to park-level greenness and non-impervious surface with areal percentages of park and open space, as well as walking distance to parks, and demographic variables like age, household income, diversity, affluence and race.
Somewhat surprisingly, initial results show no appreciable correlation between environmental facts on the ground and social factors. Explanation must now be sought in historical contingency — both in cities’ planning processes and external events. Textual analysis of cities’ comprehensive plans has been undertaken, and is helping us develop hypotheses to investigate further. Some historical investigation has begun as well, with a study of Los Angeles park planning history. Both of these are outside this paper’s scope, but will be discussed in the delivered paper.
Finally, we are also designing a human-subject experiment that will attempt to validate our measures of urban nature on the basis of human reactions to traversing through disparate “Naturehoods.” That is, to learn whether this carefully crafted statistical profile of earth surface characteristics joined with demographic statistics in fact corresponds with peoples’ “sense of place.” Living in an area dominated by strip malls and new tracts is not like living in an area with leafy boulevards, nor a downtown with a few sprinkled mini-parks, nor a wealthy enclave with large yards dense with foliage. Can digital methods predict residents’ affect in this case?
Preliminary results of these attempts at “Computing Place” appear in an interactive scholarly work  — a mapping and visualization application that allows researchers and the general public to explore will ultimately tell a story of the variation in people’s experience of nature in US cities.
1. There are of course many understandings of the term Place; my own are influenced heavily by Massey (2005) and Tuan (1977).
2. undertaken with colleagues Jon Christensen, Elijah Meeks and Maria Santos at Stanford