New York is the host city to this year’s AAG Annual Meeting. For my plenary presentation at the Population Specialty Group session I therefore decided to add a little bit of a local touch to the talk by including a new map of New York City in the slides. The work on the map came as I saw a colleague’s recent work on the population density of New York City, who also describes the data sets that I deployed in my map. The technique used for the map follows the principles of the gridded population cartogram for London.
The gridded population cartogram of New York City is a first step towards a more detailed approach to using cartograms as an alternative mapping solution for larger scale urban areas. The following map was created with the help of GIS data from the NYC Department of City Planning and small-area census data of the most recent census in 2010. The data then needed some additional processing to be suitable for a gridded cartogram transformation as described in my PhD thesis. The first version of this map uses a simple redistribution of population over a grid of 500×500 feet resolution. To create the map as a proof-of-concept, the census tracts provided sufficient accuracy to match that level of accuracy in most cases. Most census areas contained around 4-6 grid cells over which the population distribution was split in equal parts. Park and water areas without census data were embedded in the grid and were assigned 0 population values. For further improvements on the grid additional efforts would be desirable to improve on the accuracy and to eliminate errors that may result from the even redistribution, or even obtain a higher-resolution grid that allows for more detailed insights.
The population grid was then transformed using a diffusion-based method for creating density-equalising cartograms, with the population value being the determining value of the transformation, a method which is deployed in most maps on this website to date. The resulting gridded cartogram therefore is an equal-population projection in which each area on the map represents the same number of people. The transformed grid cells indicate, to what extent the number of people varies: The smaller a grid cell becomes, the fewer people live in that space, and the larger a grid cell becomes, the higher the population density is.
The following map shows the human shape of New York City, with each of the boroughs coloured differently for a better orientation. Although more detail such as subway lines or other key features may be necessary to fully read and understand the underlying geography, the variation of the grid in its simple form demonstrates the changing patterns of higher- and lower density areas of population, such as the disappearing Central Park in the middle of Manhattan, or the smaller because less-densely populated area of Staten Island:
This map provides the base for drawing a new geography of a diverse ‘melting pot‘ such as New York City. The human shape of New York puts in the center of the map, those elements that shape and form the real geography of a city.