An Alternative Weight Matrix for the Direct Leveling Adjustment

Abstract

It is common sense to use the chi-square test of the residue´s quadratic form to analyse the discrepancies between the a priori unit of weight variance and the a posteriori one in the Least Square Method adjustment for direct altimetric geodetic surveys. The rejection of the basic hypothesis point to problems in the adjustment than compromise the reliability of the work. Different reasons may to contribute to this rejection. Among those, the present paper analyses the observations weighting, on the weight matrix, because some difficulty level or subjective wraped on this making. This analisys is supported by the conceptual and mathematical revisions and still by the execution of a practice experiment, make on representative field conditions of this type of surveys. It was used the Least Square Method in the parametric form. The results indicates alternative ways to make the weight matrix, and not by the classical form than use the inverse of leveling distances, can be more appropriate.