Abstract. The recent earthquakes in L'Aquila (Italy) and in Japan have dramatically emphasized the problem of natural disasters and their correct forecasting. One of the aims of the research community is to find a possible and reliable forecasting method, considering all the available technologies and tools. Starting from the recently developed research concerning this topic and considering that the number of GPS reference stations around the world is continuously increasing, this study is an attempt to investigate whether it is possible to use GPS data in order to enhance earthquake forecasting. In some cases, ionospheric activity level increases just before to an earthquake event and shows a different behaviour 5-10 days before the event, when the seismic event has a magnitude greater than 4-4.5 degrees. Considering the GPS data from the reference stations located around the L'Aquila area (Italy), an analysis of the daily variations of the ionospheric signal delay has been carried out in order to evaluate a possible correlation between seismic events and unexpected variations of ionospheric activities. Many different scenarios have been tested, in particular considering the elevation angles, the visibility lengths and the time of day (morning, afternoon or night) of the satellites. In this paper, the contribution of the ionospheric impact has been shown: a realistic correlation between ionospheric delay and earthquake can be seen about one week before the seismic event.