ProjNet Nad 27 to WGS 1984 Results Incorrect

Question

I'm fairly new to the ProjNet library and GIS in general so please bear with me.

I have a set of coordinates from an EPSG:32038 system and I'm trying to convert to decimal Latitude and Longitude (WGS 1984). The result from ProjNet is about 700 feet south of the supposed location.

Input coordinates:

x = 1979302.949; y = 437337.980;

WKT used to define EPSG: 32038

PROJCS["NAD27 / Texas North Central",GEOGCS["GCS_North_American_1927",DATUM["D_North_American_1927",SPHEROID["Clarke_1866",6378206.4,294.9786982138982]],PRIMEM["Greenwich",0],UNIT["Degree",0.017453292519943295]],PROJECTION["Lambert_Conformal_Conic"],PARAMETER["standard_parallel_1",32.13333333333333],PARAMETER["standard_parallel_2",33.96666666666667],PARAMETER["latitude_of_origin",31.66666666666667],PARAMETER["central_meridian",-97.5],PARAMETER["false_easting",2000000],PARAMETER["false_northing",0],UNIT["Foot_US",0.30480060960121924]]

The result from ProjNet are:

latitude    32.866837297179472  
longitude   -97.56741127050887

I cross checked with NOAA's SPC to GEODETIC online tool and the results converted from DMS to decimal Lat Long are:

latitude    32.86879
longitude   -97.567574

The result from NOAA is a lot closer than what I'm getting from ProjNet. The WKT is copied from Spatial Reference website and cross checked with ESPG website. At this point, I'm not sure if the WKT used is incorrect or an issue in ProjNet.

Answer 1

I think you can't compare both results. Projection conversions in a geodetic enviroment (and NOAA's conversion suppose) are calculated using more parameters than the WKT definition. (Like points measured in both Reference Systems).

Answer 2

You are correct... ProjNET and its derivatives (like ProjNET4GeoAPI) can not transform NAD27 to NAD83 accurately. The same is true for other non-ellipsoidal conversions such as Australia's AGD84 To GDA94, and New Zealand's NZGD1949 to NZGD2000.

I have also yet to solve this problem. However it is evident that you need to convert them with a transformation grid.

The international standard for the transformation grids are called NTV2 files, and the files come in both binary (.gsb) and ascii text (.gsa) forms.

To do the transformation, you can use the routines in the "libntv2.c" file, which is available from GitHub ("https://github.com/Esri/ntv2-file-routines/tree/master"). The GitHub collection also includes a sample program ("ntv2_cvt.exe") that is a simple command line application that uses "libntv2.c" for transformations.

A collection of NTV2 files for various coordinate transformations is available at "https://www.killetsoft.de/t_ntv2_e.htm#nordamerikanischerkontinent". But unfortunately, at the time of this writing, this site only includes a NAD27 to NAD83 transformation grid for Canada which does not include all of the USA. Using the file "NTV2_0.gsb" from this website with the "libntv2.c" routine "ntv2_transform" works fine for Canada. I have compared the results of my transformation with the online tool "https://webapp.csrs-scrs.nrcan-rncan.gc.ca/geod/tools-outils/ntv2.php" and they match (but again, the tool is also Canada only).

But without a similar transformation grid for the USA, I am stumped.

The website "https://support.esri.com/en-us/knowledge-base/faq-what-national-transformation-version-2-ntv2-geograp-000003274" also does not list a suitable transformation grid for the USA.