Different Gravimetric Geoid Models over Iran with and without Terrestrial Gravity Data

The quantity and quality of the terrestrial gravity plays a major role in the accuracy of local geoids. In countries, like Iran, where the gravity database is sparse and the quality is variable, the right selection of the computational technique is a key issue. Taken all limitations into account this study discusses the computation of the Iranian geoid. The collected gravimetric database consists of 16806 gravity points corresponding to 98 points per square km on average. This gravity database involves both old (low accuracy) and new, higher accuracy gravity. In this study the authors investigated two different procedures, one uses only geopotential models, corrected for the topographic and atmospheric effects and in the second group they also applied the different modification techniques of the classical Stokes’s formula to integrate the terrestrial gravity data. The different gravimetric geoid solutions have been compared to GPS/levelling derived geoidal heights at some 200 points. The best gravimetric geoid solution has shown -117 cm mean bias and ±67 cm standard deviation with respect to the GPS-levelling geoid. The same GPS-levelling comparison has resulted -25 cm mean offset and ±114 cm standard deviation for the existing official Iranian geoid. After applying a (linear) fit to the residuals the standard deviation decreased to ±36 cm. As expected the solution without terrestrial data has shown much worse statistics, so it has no practical advantage.