Where can I find the land areas of census block groups? (AND another question)

[platypeanArchcow]

I've been searching the census website for a couple hours and I can't find this data.  (They do have a nice website for downloading pretty much any other census-related data, http://factfinder.census.gov/servlet/DownloadDatasetServlet?_lang=en.)  I found a website that displays the area of a single block group when I ask it to (like this) so the data must be somewhere, but I can't find a way I can download all the land areas of census block groups (or any other geographical areas for that matter) for, say, an entire county.

Does anyone know where to find this?  Or, say, has it and can email me a copy?

[muon2]

The data for geographic areas is in the TIGER files, which are designed for use in a GIS program. The geography page links to different files that describe the blocks and tracts. Here's a quote from a page that describes the format:

You must be logged in to read this quote.

[jimrtex]

I've been searching the census website for a couple hours and I can't find this data.  (They do have a nice website for downloading pretty much any other census-related data, http://factfinder.census.gov/servlet/DownloadDatasetServlet?_lang=en.)  I found a website that displays the area of a single block group when I ask it to (like this) so the data must be somewhere, but I can't find a way I can download all the land areas of census block groups (or any other geographical areas for that matter) for, say, an entire county.

Does anyone know where to find this?  Or, say, has it and can email me a copy?

http://www.census.gov/geo/www/cob/bdy_files.html

Select an area type, select Shapefile, select a state, download the zipfile.  The zipfile includes the shapefile (.shp) and a DBase file (.dbf).  The dbf file includes areas.

OR

Use Census Bureau American Fact Finder.  Select 2000 Census.  Select Geo within Geo.  Select for example Block Groups within State (if you are doing California, you might not be able to do all BG in the state on a single go.  I was able to do New Jersey because it had 6510 block groups.

Select "All Block Groups", click on Next, click on attributes that you want (for example total population),  and display.  Under Options, select "show geographic attributes", among the many attributes are the land area.

Click on Download and select a format.  I generally use tab delimited txt with rows and columns transposed (so you will have one row per block groups).  I don't have Excel, and if I read an .xls files with my spreadsheet program, I sometimes get some stuff that my program doesn't understand, so I end up copying from the xls file to my spreadsheet format.  I find it easier to deal with the raw data in the txt file.

[bgwah]

An easier way if you're looking for a quick solution would be to just use a little math... If you have the population and the population density, just divide the population by the density and you should get the land area.

Here is the city of Seattle, for example:
Population:  617,334 people
Density: 7,361/sq m

617,334 / 7,361 = 84, which is the land area of Seattle in square miles. You can do this with block groups and census tracts pretty easily to get an approximate land area.

[platypeanArchcow]

Thanks, Jim!

Yeah, bgwah, I would've done that, but the census tables don't include density either, at least for block groups.

[platypeanArchcow]

Another question: can I find a data file that will tell me which census-defined "urban area", if any, contains a given block group (or block, if it has to be that)?  Again, I just can't find this.  It doesn't seem to be contained in any of the geography files; the census website just gives shapefiles and maps of urban areas.

[bgwah]

Another question: can I find a data file that will tell me which census-defined "urban area", if any, contains a given block group (or block, if it has to be that)?  Again, I just can't find this.  It doesn't seem to be contained in any of the geography files; the census website just gives shapefiles and maps of urban areas.

Census tracts can sometimes be split.

Either the thematic or reference map features should show you what you want to know... Unfortunately I don't know of a faster/easier way to get the info.

[jimrtex]

Another question: can I find a data file that will tell me which census-defined "urban area", if any, contains a given block group (or block, if it has to be that)?  Again, I just can't find this.  It doesn't seem to be contained in any of the geography files; the census website just gives shapefiles and maps of urban areas.

Under the American Fact Finder, you can select using "Geo within Geo" block groups within a particular urban area.  Also among the geographic identifiers are which urban area if any that a block group or block belongs to.

Urban areas are areas of continuous high density population.  They can be defined using either block groups or census blocks.  A block group is comprised of census blocks and typically has a population around 1000.  Most census blocks have 0 population, and thus 0 density, and even in cities there may be non-populated blocks, so it may be difficult to define the limits of high density.  So block groups are used as the first part of the definition.  But block groups may contain considerable areas that are non-residential, so that their density is not representative of the residential areas.  So in those cases, the census bureau goes back to using census blocks.  So a particular urban area is comprised of census blocks, but those census blocks may comprise the entirety of some urban areas.

Just curious, what are you trying to do?

[platypeanArchcow]

Thanks!

I'm trying to visualize density profiles: y% of people in this urban area live at density at least x.  I have to run to class in a bit, I might write more later though.

[jimrtex]

Thanks!

I'm trying to visualize density profiles: y% of people in this urban area live at density at least x.  I have to run to class in a bit, I might write more later though.

It sounds like you may be getting into the same sort of problem that the Census Bureau does in defining the urban areas.  One of the population items is urban and rural population.  I looked at the block groups totally or partially within the Abbeville, LA urban cluster.  Of the 19 BG, 11 were entirely within the UC, and 8 were partially.  But there significant chunks of urban and rural population in the 8 partial.  One had 57 persons urban, and 1408 rural, so there was presumably a few blocks or block that was within the BG, adjacent to the city.  In other cases, about 1/2 the population of the BG was in the urban cluster.

The area of the mixed urban-rural BGs was significantly higher, so even if you included the rural population they would be significantly lower.  But it would be a lot of work to include the urban census blocks, which likely have similar density.