The geometric correction of remote sensing images becomes a key issue in production and updating digital maps, multisource data integration, management and analysis for many geomatic applications. 2D polynomial functions are the most prevalent to achieve this correction. Previous researches have shown that the application of 2D polynomials is conditioned by the planarity of the terrain and the uniform distribution of ground control points, but did not explicitly discuss the criteria for evaluating the success or failure of their application. In this study, we will try to give some of these criteria and to develop some old analog cartographic rules to suit the nature of the digital satellite images. In this research, we discussed mathematical foundation for evaluating the precision of control points- based geometric correction of satellite images. We have also tested the effect of the topography of the imaged scene on this accuracy. The test has been carried out by the use of satellite images extracted from Google Earth. These images cover some areas in the city of Latakia in Syria. Also, control points have been extracted from Google Earth and transformed into the Syrian stereographic coordinates system. Results demonstrated that the second degree 2D polynomial is very suitable for plan small scenes with uniform distribution of the control points over the entire scene.