Frame mesh

In geodesy , the framework network method is used for extensive network adjustment when the matrix of normal equations becomes too large for stable numerical inversion - i.e. H. when the survey network has too many points for an exact solution.

It is an approximate or iteration method, in which a frame is first defined from a dozen to a few hundred dot meshes, which is then subjected to strict adjustment. All the other measurement points are subsequently integrated into this framework, which is interlocked as closely as possible .

The individual point meshes are preferably defined at those points - mostly in the vicinity of large cities - where many measuring lines or triangular chains of triangulation converge, so that the frame results in a geometrically and numerically stable structure.

This made it possible in the first quarter of the 20th century to precisely calculate geodetic networks over large parts of a continent , although computing aids such as today's EDP ​​were not available.

The two most popular uses are:

• the North American network NAD , the traverses of which - broken down into large squares according to the four cardinal points - were leveled around 1925 using the Bowie method . It covers around four million km² with 26  nodal networks and a normal equation matrix of 100 × 100, which at that time was just possible manually. The almost 10,000 individual TPs were then fitted into this frame with an average mesh size of 500 km  .
• the Central European Network , which began around 1940 and was completed after the Second World War under the leadership of Helmut Wolf and officers of the US occupation forces. It formed the basis of the later European network (see also  ED50 and  ED79 ), which, however, was calculated using "seams" along the national borders so that the national surveys of the individual states did not have to be left to each other in full.