GPS as a height measuring tool has been widely introduced for solving various tasks within the fields of Geodesy and Engineering Surveying. The obvious advantage of not needing lines of sight when acquiring height differences is, however, darkened by the fact that GPS and traditional spirit levelling have fundamentally different reference systems for heights. To handle the main task, i.e. to replace spirit levelling with GPS levelling, or other important tasks, like, for instance, detection of vertical crustal movements or geoid slope and undulation contribution for creating geoid models, knowledge about the two reference systems, and how to combine them, is required.

In this thesis a practical experiment concerning GPS as a height measuring tool is described. The experiment, whos aim is to develop and evaluate a geometric geoid model, has been carried out in an area of approximately 100x300 km situated in central Sweden, were GPS measurements (with average baseline length of 13 km) have been conducted at well established benchmarks with known orthometric heights.

In order to gain experience, valuable for the experiment, GPS measurements were carried out on one and the same line (referred to as base measurements) during different circumstances, i.e. day, night, summer, winter etc., during the course of a year. The findings lead to strategies which were applied in the field work as well as in the post processing procedures of field collected data.

Different gravimetric geoid models have been evaluated on a regional bases utilising the Swedish GPS reference network SWEPOS and locally, in the research area, utilising the results from the GPS levelling.

To find out if a geometric geoid model, based on the GPS measurements, would be a better height corrector surface for the research area than the gravimetric geoid models, such a model has been computed with geostatistical methods, i.e. universal kriging. The model has been experimentally evaluated utilising a Swedish GPS campaign called RIX95. It was found, when the geometric geoid model is used, that the absolute accuracy of GPS levelling is ±14 mm and the relative accuracy ±10 mm (on a 10 km long baseline). It is concludet that the geometric geoid model is the best height corrector surface for the research area among the studied models.