Dmitry: When you use the GIS shape file from http://www.census.gov/geo/www/cob/bdy_files.html you<http://www.census.gov/geo/www/cob/bdy_files.html%20you> are using a "generalized GIS file." The limitations of these files are listed on the CB website as follows: "The generalized files have a much smaller file size than the original file extraction from the Census Bureau's TIGER database, resulting in faster download and processing times. Limitations Because of coordinate thinning: 1. Cartographic boundary files should not be used for geocoding; 2. Some offshore, redundant, zero population and housing land areas may be absent from the files; 3. Cartographic Boundary files are not necessarily vertically integrated with previous boundary file sets." For smaller geographies such as TAZs, you are better off using a detailed shape file/Any other GIS file derived from TIGER directly. Thanks Nanda From: ctpp-news-bounces(a)chrispy.net [mailto:ctpp-news-bounces@chrispy.net] On Behalf Of Messen, Dmitry Sent: Wednesday, February 22, 2012 8:45 PM To: ctpp-news(a)chrispy.net Subject: [CTPP] 2000 Census Traffic Analysis Zones Would anybody know what the deal was with the 2000 Census Traffic Analysis Zones? I am working with CTPP 2000 Table 3 data. To do some spatial analysis, I turned to census boundary files for traffic analysis zones (http://www.census.gov/geo/www/cob/bdy_files.html). I quickly realized that 63 out 2639 zones for the Houston region are represented by 2+ non-adjacent polygons. Does this happen in other regions as well? Was this delineation done purposefully or perhaps these are simply errors stemming from TIGERLine 2000? Any input will be much appreciated. Thank you. Dmitry Messen H-GAC dmessen@h-gac.com<mailto:dmessen@h-gac.com>

Dmitry, It has been an entire decade since I worked on the TAZ geography here in Minneapolis-St. Paul, so my memory is not as sharp on the subject. The software used in 'TAZUP' at the time was not as thorough, so to speak, as that used this time around. Visually scanning the map that was on the computer screen before you often required 'zooming in' to locate missing polygons; they were often rather difficult to spot. I do not recall that the software had any 'checks and balances' like it did for the 2010 work effort. Consequently, the situation you indicated could quite possibly have taken place. I believe that I had a few instances like you described. Fortunately, they were tiny slivers. Good luck with the work! Bob Paddock Transportation Planning Metropolitan Council Bob.paddock(a)metc.state.mn.us 651 / 602-1340 From: ctpp-news-bounces(a)chrispy.net [mailto:ctpp-news-bounces@chrispy.net] On Behalf Of Messen, Dmitry Sent: Thursday, February 23, 2012 8:57 AM To: &#39;ctpp-news(a)chrispy.net&#39; Subject: Re: [CTPP] 2000 Census Traffic Analysis Zones Nanda, Yes, I know about the generalization; however, it is my understanding that it results in the simplification of the lines and shouldn't alter the polygon topology. Nevertheless, we'll go ahead and process the TIGERLine files. But the question remains: did the Census standards for Traffic Analysis Zones in 2000 geography specifically allow non-contiguous zones (other than the islands, of course)? From: ctpp-news-bounces(a)chrispy.net [mailto:ctpp-news-bounces@chrispy.net] On Behalf Of Srinivasan, Nanda Sent: Thursday, February 23, 2012 8:03 AM To: ctpp-news(a)chrispy.net Subject: Re: [CTPP] 2000 Census Traffic Analysis Zones Dmitry: When you use the GIS shape file from http://www.census.gov/geo/www/cob/bdy_files.html you<http://www.census.gov/geo/www/cob/bdy_files.html%20you> are using a "generalized GIS file." The limitations of these files are listed on the CB website as follows: "The generalized files have a much smaller file size than the original file extraction from the Census Bureau's TIGER database, resulting in faster download and processing times. Limitations Because of coordinate thinning: 1. Cartographic boundary files should not be used for geocoding; 2. Some offshore, redundant, zero population and housing land areas may be absent from the files; 3. Cartographic Boundary files are not necessarily vertically integrated with previous boundary file sets." For smaller geographies such as TAZs, you are better off using a detailed shape file/Any other GIS file derived from TIGER directly. Thanks Nanda From: ctpp-news-bounces(a)chrispy.net [mailto:ctpp-news-bounces@chrispy.net] On Behalf Of Messen, Dmitry Sent: Wednesday, February 22, 2012 8:45 PM To: ctpp-news(a)chrispy.net Subject: [CTPP] 2000 Census Traffic Analysis Zones Would anybody know what the deal was with the 2000 Census Traffic Analysis Zones? I am working with CTPP 2000 Table 3 data. To do some spatial analysis, I turned to census boundary files for traffic analysis zones (http://www.census.gov/geo/www/cob/bdy_files.html). I quickly realized that 63 out 2639 zones for the Houston region are represented by 2+ non-adjacent polygons. Does this happen in other regions as well? Was this delineation done purposefully or perhaps these are simply errors stemming from TIGERLine 2000? Any input will be much appreciated. Thank you. Dmitry Messen H-GAC dmessen@h-gac.com<mailto:dmessen@h-gac.com>

To answer your question specifically, I think the multi-TAZ area I found did include non-contiguous areas, but I would have to go back to be certain. Karen Karen M. Lorenzini, P.E., AICP Texas Transportation Institute 512/467-0952, x-12121 k-lorenzini(a)ttimail.tamu.edu From: Lorenzini, Karen Sent: Thursday, February 23, 2012 10:23 AM To: &#39;ctpp-news(a)chrispy.net&#39; Subject: RE: 2000 Census Traffic Analysis Zones Hi, Dmitry, I had an experience similar to the one you describe. CTPP 2000 TAZ data was showing over 1000 population and over 300 households for a zone that was still entirely undeveloped in 2010 Googlemaps imagery. I couldn't figure it out until I imported the CTPP TAZ geography and when I tapped on that TAZ: a huge, multi-TAZ area surrounding the MPO boundary lit up. It appeared that population and households in those surrounding areas had been included in the TAZ inside the MPO boundary. Census STF1 block group data was correct. This was a different Texas MPO area than H-GAC covers. Karen Karen M. Lorenzini, P.E., AICP Texas Transportation Institute 512/467-0952, x-12121 k-lorenzini@ttimail.tamu.edu<mailto:k-lorenzini@ttimail.tamu.edu> From: ctpp-news-bounces@chrispy.net<mailto:ctpp-news-bounces@chrispy.net> [mailto:ctpp-news-bounces@chrispy.net] On Behalf Of Messen, Dmitry Sent: Thursday, February 23, 2012 8:57 AM To: &#39;ctpp-news(a)chrispy.net&#39; Subject: Re: [CTPP] 2000 Census Traffic Analysis Zones Nanda, Yes, I know about the generalization; however, it is my understanding that it results in the simplification of the lines and shouldn't alter the polygon topology. Nevertheless, we'll go ahead and process the TIGERLine files. But the question remains: did the Census standards for Traffic Analysis Zones in 2000 geography specifically allow non-contiguous zones (other than the islands, of course)? From: ctpp-news-bounces@chrispy.net<mailto:ctpp-news-bounces@chrispy.net> [mailto:ctpp-news-bounces@chrispy.net] On Behalf Of Srinivasan, Nanda Sent: Thursday, February 23, 2012 8:03 AM To: ctpp-news@chrispy.net<mailto:ctpp-news@chrispy.net> Subject: Re: [CTPP] 2000 Census Traffic Analysis Zones Dmitry: When you use the GIS shape file from http://www.census.gov/geo/www/cob/bdy_files.html you<http://www.census.gov/geo/www/cob/bdy_files.html%20you> are using a "generalized GIS file." The limitations of these files are listed on the CB website as follows: "The generalized files have a much smaller file size than the original file extraction from the Census Bureau's TIGER database, resulting in faster download and processing times. Limitations Because of coordinate thinning: 1. Cartographic boundary files should not be used for geocoding; 2. Some offshore, redundant, zero population and housing land areas may be absent from the files; 3. Cartographic Boundary files are not necessarily vertically integrated with previous boundary file sets." For smaller geographies such as TAZs, you are better off using a detailed shape file/Any other GIS file derived from TIGER directly. Thanks Nanda From: ctpp-news-bounces@chrispy.net<mailto:ctpp-news-bounces@chrispy.net> [mailto:ctpp-news-bounces@chrispy.net] On Behalf Of Messen, Dmitry Sent: Wednesday, February 22, 2012 8:45 PM To: ctpp-news@chrispy.net<mailto:ctpp-news@chrispy.net> Subject: [CTPP] 2000 Census Traffic Analysis Zones Would anybody know what the deal was with the 2000 Census Traffic Analysis Zones? I am working with CTPP 2000 Table 3 data. To do some spatial analysis, I turned to census boundary files for traffic analysis zones (http://www.census.gov/geo/www/cob/bdy_files.html). I quickly realized that 63 out 2639 zones for the Houston region are represented by 2+ non-adjacent polygons. Does this happen in other regions as well? Was this delineation done purposefully or perhaps these are simply errors stemming from TIGERLine 2000? Any input will be much appreciated. Thank you. Dmitry Messen H-GAC dmessen@h-gac.com<mailto:dmessen@h-gac.com>