The NGS has issued a notice on their website regarding observed discrepancies between the 2009 and 2012 GEOID models. The notice expansion on August 8, 2012, came in the wake of analysis performed by the LSU Center 4 GeoInformatics back in the week of July 15th where we identified issues in Louisiana, Mississippi, Texas, and other states within the CONUS. C4G notified the NGS of the issues found during the ESRI conference in San Diego, CA during the week of July 25th and we are working with them to resolve the problems as soon as possible.
You may find it useful to do your own testing of GEOID12 in your project area but C4G is recommending that everyone in Louisiana stick with GEOID09 until we learn more about what caused the issues we are seeing in GEOID12 and have confidence that NGS has resolved or justified the differences we are seeing to our satisfaction.
You can keep up with this and other C4G related topics by following us on Twitter, Facebook, YouTube, C4Gnet RSS feed and the C4G website. The following is a copy of the NGS Notice Update.
**NOTICE UPDATE**
GEOID12 errors found in northern Gulf Coast region
The National Geodetic Survey (NGS) has determined that the new hybrid geoid model, GEOID12, contains localized errors in Louisiana, Mississippi, Texas, and Alabama. Some of these errors are in addition to those identified in the previous notice. Click here for more information.
Vertical controls for all surveys shall be determined in the North American Vertical Datum of 1988 (NAVD88). All measurements shall be referenced to local control stations of the National Spatial Reference System, specifically the public domain Louisiana State University Continuously Operating Reference Stations network or other currently National Oceanographic and Atmospheric Administration National Geodetic Survey approved reference stations, such as benchmarks, monuments, or continually operating reference stations.
The National Geodetic Survey (NGS) issued a notice on Saturday, August 4, 2012 stating that they have determined that the new hybrid geoid model, GEOID12, contains localized errors in Louisiana and Mississippi. Click here for more information.
GPS only provides ellipsoid height – very accurate and very precise, but that is not elevation. To obtain elevation information from an ellipsoid height we must utilize a theoretical model of the earth’s gravity field called “the geoid.” Our knowledge of the geoid is only approximate, and it is very faulty in Louisiana because of the constant surface and subterranean movement due to consolidation, faulting, salt domes, and subsidence – the result being largely responsible for “coastal erosion” and the inability of benchmarks to remain reliable for more than a couple of years or so. Therefore, the primary thrust of research must be with regard to a better understanding of the geoid as it exists in Louisiana coupled with the systematic processing of the decade of GPS observations archived in LSU’s server banks. 
We have a research plan that is in physical geodesy which involves GPS observations, coupled with absolute gravity field observations at the same locations, and with zenith camera observations to determine the slope of the geoid at those same locations. Eventually, many of those locations will need to be observed in precise geodetic leveling campaigns (a very expensive proposition also).
In the 1980s, benchmark densification projects were undertaken in Orleans, Jefferson, St. Bernard, and Plaquemines Parishes. These projects consisted of 1st- and 2nd-Order Geodetic Leveling observations accompanied by 2nd-Order Relative Gravity (double ladder) observation surveys. The total number of benchmarks surveyed was over 500 monuments, and all observations and descriptions were Blue Booked, accepted, and published by the National Geodetic Survey. This was later followed by FG-5 Absolute Gravity observations first at the University of New Orleans which was observed multiple times and then at numerous GPS Continuously Operating Reference Station (CORS) sites throughout the State of Louisiana, some more than once by either the National Geodetic Survey or the National Geospatial-Intelligence Agency. The density of these observations were at approx. 100 mile distances, and are not sufficient to provide significant data points to improve the geoid model. The primary purpose for these observations was to provide an independent check on changes in ellipsoid heights observed at the LSU CORS sites.
Equipment needed is an A-10 Absolute Gravity Meter and a CODIAC Zenith Camera to supplement GPS observations in South Louisiana (and eventually North Louisiana). The National Geodetic Survey recommends an observational density grid of 40 kilometers which will occupy field researchers for many years. That data combined with our existing database will insure that the LSU GPS Real Time Network will provide centimeter-level accuracy for elevations throughout the state; a prime example of the direct impact that research has on practical applications.
NOTICE: NGS Update, August 17, 2012
The National Geodetic Survey has determined that the GEOID12 model was developed using erroneous input control points, primarily in the Gulf Coast region. Click here for more information. NGS immediately corrected these errors and developed the GEOID12A replacement model. GEOID12A is now available for public analysis and comment on the NGS Beta website for GEOID12A and will soon be released for production.
C4G will analyze GEOID12A and publish our findings as soon as possible. Click here to see a map of GEOID12A minus GEOID12, changes in the data used for GEOID12A are also noted on this map.
You can keep up with this and other C4G related topics by following us on Twitter, Facebook, YouTube, C4Gnet RSS feed and the C4G website.
