Submitted by Steve (not verified) on Sun, 2006-07-30 04:24.
Hey Dylan:
If you look at this page
http://edndoc.esri.com/arcobjects/9.1/default.asp?url=/arcobjects/9.1/ComponentHelp/esriGeometry/esriSRGeoTransformation3Type.htm
You can see the description as Quebec. I think this means that the particular NAD conversion is tweaked for Quebec. If you look at some of the other constants you will see that there are other NAD to NAD custom conversions that must be tweaked for certain areas. Don't ask me why ESRI chose that one as the default. This is all of course conjecture. The only way to nail this down is to look at the parameters that it uses for the NAD shift. I thought it might be hidden somewhere in the file system but I don't see it.
Check out this doc
http://www.usace.army.mil/usace-docs/eng-manuals/em1110-2-1003/c-5.pdf
and search for Quebec. The later results show that there is a lot of crustal movement about Quebec and the Great Lakes so perhaps this is a "realignment" of the transformation which takes these changes into account.
By the way, thanks for doing this benchmarking. People do this stuff all the time with Statistical software as well. They usually use SAS as the gold standard. It was based upon some of these results:
http://www.practicalstats.com/Pages/excelstats.html
http://www.agresearch.cri.nz/Science/Statistics/exceluse1.htm
http://www.cof.orst.edu/net/software/excel/no-stats.php
http://www-unix.oit.umass.edu/~evagold/excel.html
that I advise people to NEVER do statistical analysis in Excel except for things like Mean and Median.
Some information
Hey Dylan:
If you look at this page
http://edndoc.esri.com/arcobjects/9.1/default.asp?url=/arcobjects/9.1/ComponentHelp/esriGeometry/esriSRGeoTransformation3Type.htm
You can see the description as Quebec. I think this means that the particular NAD conversion is tweaked for Quebec. If you look at some of the other constants you will see that there are other NAD to NAD custom conversions that must be tweaked for certain areas. Don't ask me why ESRI chose that one as the default. This is all of course conjecture. The only way to nail this down is to look at the parameters that it uses for the NAD shift. I thought it might be hidden somewhere in the file system but I don't see it.
Check out this doc
http://www.usace.army.mil/usace-docs/eng-manuals/em1110-2-1003/c-5.pdf
and search for Quebec. The later results show that there is a lot of crustal movement about Quebec and the Great Lakes so perhaps this is a "realignment" of the transformation which takes these changes into account.
By the way, thanks for doing this benchmarking. People do this stuff all the time with Statistical software as well. They usually use SAS as the gold standard. It was based upon some of these results:
http://www.practicalstats.com/Pages/excelstats.html
http://www.agresearch.cri.nz/Science/Statistics/exceluse1.htm
http://www.cof.orst.edu/net/software/excel/no-stats.php
http://www-unix.oit.umass.edu/~evagold/excel.html
that I advise people to NEVER do statistical analysis in Excel except for things like Mean and Median.