My first abstract...

The goal of this study was to analyse the possibility to determine the orthometric height with GPS, along with a geoid model, instead of using geometric levelling in the Porto region. To achieve this goal we used a recently observed levelling line between Porto and Crestuma–Lever dam and GPS and gravimetry observations at 5 points belonging to this line.
We concluded that the influence of the gravimetric corrections on this levelling line was not significant due to the small height variations of approximately 70m. This allowed us to use the geometric levelling directly as a geopotential level that could be used as a reference to evaluate the use of GPS minus a geoid undulation model in the calculation of orthometric heights in this area.
The geoid undulation values were calculated from the model computed by Catalão [2007]. 5 points from line of levelling were selected and GPS and gravity data were collected at these locations. In order to ensure redundancy in the observations, these points were observed during three campaigns with a vertical repeatability up to 30 mm In addition, the influence of the chosen method of interpolation of geoid model can affect results by up to 40 mm. This gives a combined error for the observation procedure of about 50 mm. Since the difference between GPS minus geoid model versus levelling reached 100 mm, we conclude that the intrinsic model error can also reach 50 mm.
It may be concluded that the sum of the uncertainty in the GPS observations, the geoid interpolation methods and the geóide model error itself does not permit, yet, to obtain values under to decimetre in the calculation of orthometric height using GPS and a geoid model.