While revolutionary in 1929, the old way of doing things has been improved with research in geodesy as well as the fact that some areas of the country’s topography have changed over time. NGVD29 was based on what is called a “constrained least-squares network adjustment.” Since that original system was devised, our knowledge of the Earth’s crust has increased such that we can now measure tiny incremental changes in our system of elevation benchmarks. New surveys have become increasingly difficult to “fit” into the old system without warping the new observations as they are forced to be compatible with the old values. A study by the National Academy of Sciences found that the old system of NGVD29 had deteriorated to become an incompatible reference system and that the North American Continent needed to start all over again with something better. The new North American Vertical Datum of 1988 is a great improvement over its predecessor in terms of mathematical techniques employed as well as being based on the Earth’s observed gravity field. NAVD88 is compatible in principle with the Geoid, the mathematical system of describing the relation of the Earth’s gravity field to elevations.
We must first define the two datums.
The National Geodetic Vertical Datum of 1929 (NGVD29), previously referred to as Sea Level Datum of 1929, was the network of over 20,000 miles of levels constrained to Mean Sea Level (MSL) at 26 tide stations around North America. The network was warped due to variations in the Local Mean Sea Level (LMSL) at those 26 tide stations. These variations introduced errors into the network adjustment. The datum is not mean sea level and was renamed the National Geodetic Vertical Datum of 1929 in 1973.
North American Vertical Datum of 1988 (NAVD88), completed in June of 1991 contained an additional 100,000 miles (15,000+ miles of new levels) of levels and was a minimally-constrained adjustment, constrained only to the primary tidal benchmark at Father Point/Rimouski, Quebec, Canada.
Approximately 100,000 miles of leveling had been added to the National Geodetic Reference System (NGRS) since NGVD 29 was created. In the early 1970s, NGS conducted an extensive inventory of the vertical control network. Many existing benchmarks were affected by crustal motion associated with earthquake activity, postglacial rebound (uplift), and subsidence. Other problems (distortions in the network) were caused by forcing the 100,000 miles of leveling to fit previously determined NGVD 29 height values. NAVD88 was created to eliminate those errors, incorporate the additional leveling, and to produce a new network that is consistent with both conventional and GPS leveling.