Urbanization of the United States over two centuries: an approach based on a long-term database (1790–2010)

Abstract

International audienceThis paper presents an original methodology for the construction of a harmonized database for urban areas in the United States from 1790 to 2010 (a period with a census every 10 years, amounting to 23 dates). Upstream, this method is based on a theoretical choice, the use of a spatio-temporal reference frame to construct urban areas over the long term, defined by a maximum range of one hour’s travel for each period. We used an automatic aggregation from a Reilly model. Urban measures (growth, concentration) derived from this harmonized database are then presented, providing a unified picture of urbanization in the United States over more than two centuries. They are compared with results derived from more classical databases established using the official figures from the Census Bureau, or from other constructions differing from the present system. The results diverge considerably in some instances, which confirms the impact of the method of delineation of geographical areas in measures of growth

The Old and the New: Qualifying City Systems in the World with Classical Models and New Data

“…Building on a longstanding geographical view on the hierarchical organization of urban systems (Berry ; Pred ) and its revised version as an evolutionary process (Pumain ) we consider the definition of cities as geographical entities to be necessary at mainly two spatio‐temporal scales: cities as evolutionary centers of concentration of people and activities keeping a consistent meaning at the local scale as daily urban systems (strong interactions connecting in average three or four different places of activity per person) and cities as attractive nodes within competing networks of cities at higher scales of space and time (less frequent interactions but generating interdependencies leading to interurban co‐evolution). That is the reason why, for international comparisons, we rely on standardized databases where “cities” are not taken for granted from the official statistical definitions but designed according to similar criteria (Bretagnolle et al ). These harmonized databases have been built for seven large regions of the world (Fig.…”

“…Different data sets might lead to different conclusions, as demonstrated by Bretagnolle, Giraud, and Mathian () on the United States case. Ideally, the model should be tested over long periods of time, on large number of towns and cities, using various delineations, including (variable delineation) or not (constant delineation) the spatial extensions of urban areas for each time interval (Paulus ; Bretagnolle et al ). This last test was not possible for all the data sets examined here and thus our results must be considered cautiously.…”

What do analyses of city size distributions have in common?

“…Another cluster (in red) shows a specific use of the term "time". Originating from a single research group in France, Bretagnolle et al, (2000Bretagnolle et al, ( , 2015 and Pumain et al (2015) indeed present long-term evolutions of systems of cities, reporting on their growth and S0 in supplement for a lookup table between article identifiers and bibliographic references structure of several decades. The light blue cluster gathers comparative studies who therefore make a more thorough use of the term "countries".…”

The Old and the New: Qualifying City Systems in the World with Classical Models and New Data

“…Building on a longstanding geographical view on the hierarchical organisation of urban systems (Berry, 1964, Pred, 1977 and its revised version as an evolutionary process (Pumain, 2000 and2006) we consider the definition of cities as geographical entities to be necessary at mainly two spatio-temporal scales: cities as evolutionary centres of concentration of people and activities keeping a consistent meaning at local scale as daily urban systems (strong interactions connecting in average 3 or 4 different places of activity per person) and cities as attractive nodes within competing networks of cities at higher scales of space and time (less frequent interactions but generating interdependencies leading to inter-urban co-evolution). That is why for international comparison we rely on standardized databases where "cities" are not taken for granted from the official statistical definitions but designed according to similar criteria (Bretagnolle et al, 2015). These harmonised databases have been built for seven regions of the world 1 (Figure 1).…”