SYNTHETIC HOUSEHOLD POPULATION VIEWER | RTI
Synthetic Population Viewer Information

How does the human population organize itself across the landscape? How do age, income, race, and household size vary in cities?

The Synthetic Population Viewer helps you answer those questions by providing an interactive map of a 'synthetic population'. Each dot on the map represents a single household. Across census geography the age of the head of household, household income, race of the head of household, and household size are accurately represented in the synthetic population. Households are mapped realistically to represent actual population distributions.

But these are not 'real' households and they are not mapped to 'real' household locations. Nonetheless, they represent real households in a reasonably accurate way. The viewer contains a representation of 112,596,000 households estimated to exist in the fifty states and Washington, D.C., based on the 2010 U.S. Decennial Census and the 2007-2011 5-year American Community Survey (ACS).

Using the viewer: The basic functionality of the viewer is provided in the graphic below.

What you will see: The dynamic interaction of these five variables (location, income, age, race, and household size) are most apparent in cities with dense populations. As you get familiar with the data and the viewer application, it begins to be easy to identify certain neighborhoods simply based on the patterns of the colored dots. For example, student housing near major universities are apparent because these off-campus student households are relatively densely clustered, mostly have a single or perhaps two occupants, have low income and often have a diverse racial makeup. Patterns of race and income are perhaps most apparent as you examine the U.S. population. As you explore you'll gain an appreciation for the amazing breadth and diversity of the U.S. human population.

"Quad View" vs. "Single View": By default the viewer appears in "Single View", displaying one variable at a time in a single panel. If you choose "Quad View" each of the four variables will be shown in a separate panel. Each of the four panels color-codes households based on one of four characteristics: household size (number of occupants), head of householder age, head of householder race, and household income. You can zoom and pan in any one panel and the other three panels will update accordingly. Use your mouse's scroll wheel or the +/- icons in the upper left to zoom in or out. Click and drag in any panel to pan across the map. The synthetic population households are only shown at zoom levels 10-13.

Backgrounds: The synthetic population can be viewed with a standard map background ("Map Background"), or with a black background ("Black Background"), or with a white background ("White Background"). The difference between the "Map Background" and the "Black Background" is shown below.

In Single View, large areas (depending on the size of your monitor) can be viewed. In the image above, almost all of the entire Los Angeles basin is in the view. Yellow colors indicate lower income and blue areas indicate higher income. Higher incomes tend to occur along the coast, in the canyons north of the city, and along the edges of the overall image.

Errors: Why are there dots on roads, parks, water? You will see 'errors' in the placement of the points. Some will appear in parks, cemeteries, roads, water and other places where households would not or are not located. This is because the data are a model and not a completely accurate map of households. Refinements to the spatial position of the households are planned. However, at a general scale, the overall household map is relatively accurate. Apartment buildings are not represented in the data -- each household is presented as a separate dot with a separate location. Places with high-rise apartment buildings are illustrated by a high density of household dots.

The Data: The RTI 2010 U.S. Synthetic Population Ver. 1 (Wheaton, 2014) was based on original work from Los Alamos National Laboratory (Beckman, 1996). The 2010 U.S. Synthetic Population was developed to match U.S. household characteristics from the 2010 U.S. Decennial Census and the 2007-2011 American Community Survey at the census block group level. The distribution of households is based on Integrated Climate and Land Use Scenarios (ICLUS) 90-meter population estimates.

There is no identifying information on each household and there is no intent to map actual households to actual places. The data are created from publically available sources and no private or confidential data were used to generate the data. In fact, that is the intent of developing these data -- they should reflect the U.S. population in a realistic way, without sacrificing privacy.

Viewer Technology: The viewer uses image tiles that are pre-generated using open source GeoServer (www.geoserver.org) technology and are being served from an Nginx (nginx.org/) web server. Basic JavaScript and HTML including the OpenLayers (www.openlayers.org/) api were used to develop the browser interface. Although the 2010 U.S. Synthetic Population is stored and maintained in a postGIS database, GeoServer converts the vector-based spatial data to raster tiles at each viewable scale to enhance re-drawing speed in the viewer. The background map is from MapQuest.

Acknowledgments: This viewer and the underlying data were developed under the Models of Infectious Disease Agent Study (MIDAS) grant 1-U24-GM087704 funded by National Institutes of General Medical Sciences (NIGMS). Explore the rest of the www.epimodels.org website to see how Agent-Based Models (ABMs) are being used to study the dynamics of infectious disease transmission.

The RTI Development Team: Bill Wheaton (team lead), Brandon Bergenroth (viewer development and synthetic population data management), Dahl Winters (viewer interface), Jay Rineer (synthetic population development), Justine Allpress (synthetic population development), James Cajka (schools and business data processing-not shown in viewer), Bernadette Chasteen (group quarters data processing-not shown in viewer), Diane Wagener (Principal Investigator), Phil Cooley (Co-Principal Investigator).

Citations:

Wheaton, W.D. 2014. "2010 U.S. Synthetic Population Database: Quick Start Guide". RTI International. Retrieved from 2010_synth_pop_ver1_quickstart.pdf.

Wheaton, W.D., J.C. Cajka, B.M. Chasteen, D.K. Wagener, P.C. Cooley, L. Ganapathi, D.J. Roberts, and J.L. Allpress. 2009. Synthesized population databases: A U.S. geospatial database for agent-based models. RTI Press paper available at http://www.rti.org/pubs/mr-0010-0905-wheaton.pdf.

Beckman, R.J., K.A. Baggerly, and M.D. McKay. 1996. Creating synthetic baseline populations. Annals of Transportation Research 30(6):415-429.

U.S. Environmental Protection Agency (EPA). 2010. ICLUS v1.3 User's Manual: ArcGIS Tools and Datasets for Modeling US Housing Density Growth. Global Change Research Program, National Center for Environmental Assessment, Washington, DC; EPA/600/R-09/143F.

For further information, please contact Bill Wheaton at RTI (wdw@rti.org).