Hi everyone,
I am a building scientist at a national laboratory. We want to associate the centroids of PUMAS with future weather data that we generated. To do that, we need the latitude and longitude of the centroids of PUMAS. I downloaded PUMAS shape file at IPUMS USA. I was able to use geopandas to calculate the coordinates of the centroids. But they are in an unknown form. For example, the coordinates of the centroid of Alameda County (West)–San Leandro, Alameda & Oakland (Southwest) Cities PUMA is (-2258808.793 339396.472). Could anyone tell me how to convert these coordinates to latitude and longitude?
Thanks a lot!
The IPUMS USA shapefiles use a projected coordinate system–not longitude/latitude. The system is defined precisely in the .prj file that accompanies the shapefile. There are different ways to convert from one to the other depending on your preferred software, for example you can use the Project tool in ArcGIS. Google would be the best way to find out how to do this for other software.
Alternatively, PUMA shapefiles from IPUMS NHGIS have internal latitude and longitude fields that represent the “geographic centroid” of the PUMA and can be converted to spatial points; to get the NHGIS shape files, enter “PUMA” as the geographic level, 2010 as the focal year, and click on the “GIS Files” tab of the results page.
Hi Matthew,
I selected “PUMA” as the geographic level and 2010 as the focal year, but the returned file is PUMA 2000. Do you know how to get PUMA 2010?
Thanks,
Zhaoyun
Try removing the year filter, as in the screenshot below. In NHGIS, the year filter pertains to “data release year” rather than “delineation year.” Understandably, that distinction is sometimes confusing, but the first ACS data release to use 2010 PUMAs was the 2012 ACS, so NHGIS provides GIS files for 2010 PUMAs if you filter on the year 2012 or subsequent years. (The 2010 & 2011 ACS releases used 2000 PUMAs, and 2020 PUMAs are scheduled to be used for the first time in next year’s release of 2022 ACS data.)
Hi Jonathan,
It works. Thanks a lot for your help!