Journal of Surveying Engineering

By a minor change in the weighting policy, the complicated Helmert model can be mathematically reduced to an exact equivalent of iterated almost unbiased estimation (IAUE). Such an approach is referred to as the modified Helmert method. It is shown that the Helmert method is identical with IAUE if one works directly with the covariance matrix of the observables instead of estimating the a priori variances of unit weight, as is usually done.

The aim of this study is to evaluate the use of interpolation methods, which are used for digital terrain modeling, in transformation of Global Positioning System (GPS) ellipsoid heights to local heights. To test the interpolation methods used in this research, a set of GPS data collected in northern Turkey, 300 km from Istanbul, is used. The data were transformed to a local height system using 11 different interpolation processes, classified in four types of interpolation approaches. The data set includes 109 points, whose heights are known in both GPS and local height systems. Eighteen of them are selected as reference points. The differences between the local and GPS heights at the remaining 91 points were recalculated by using the differences at the 18 reference points. The interpolated (or extrapolated) differences were compared with the known differences for 91 points.

The main design criteria in a geodetic network are precision, reliability and strength, and economy. The proper design of geodetic networks is an integral part of the geodesist's task. The main goal of this paper is to suggest a strategy for designing a geodetic network to meet high reliability and geometrical strength criteria. In this context, a real 2D deformation monitoring scheme has been selected to evaluate its design criteria. The results indicate that the perimeter of a geodetic network is relatively weaker than the middle. This means that the perimeter observations are usually weaker than those in the middle, the maximum distortions are usually located on the edge netpoints, and the maximum robustness parameters belong to the perimeter of the network. A strategy has been recommended to counter this problem. The results also showed that the robustness parameters in rotation, shear, and scale are affected by redundancy numbers. Actually, the largest robustness parameters are due to the weak observations in the network.