GISTEL is a physical model based on the calculation of the brightness, which estimates the solar energy at the ground surface level using METEOSAT numerical images. In this paper, GISTEL is applied to B2 images in order to evaluate the reflection coefficient of the ground surface including both the geographical distribution and temporal variation over Tunisian continental areas. Apart from some seasonal fluctuations, that do not exceed 1%, the reflection coefficient shows a rather regular variation from the north to the south of Tunisia. The obtained results are in good agreement with other works, as well as the ground stations measurements made by the Tunisian National Institute of Meteorology. As the reflection coefficient stands for a characteristic parameter of the ground, it is shown to be very useful in many energetic, climatic and agronomic applications, which makes it decisive in the determination of the economic policy of the country. Its presentation in the form of maps makes it more accessible to users.