Differential Global Satellite Navigation Systems (DGNSS) are based on the principle that the main sources of error in satellite navigation are consistent over substantial geographical areas. The magnitudes of these errors can be measured by installing reference receivers at fixed, known locations. The corrections they generate are then broadcast via a radio system. DGNSS users in the vicinity receive these corrections and employ them to adjust their own position measurements accordingly. Marine radiobeacons are widely used to transmit these correction messages to maritime users. These radiobeacons enjoy an existing, protected, frequency band and large numbers of them are available world-wide. Recent research has studied the many factors that affect the coverage of radiobeacons. These include propagation losses, skywave-borne interference from the many beacons that share the radio band, and atmosphe1ic noise. The results have been embodied in a widely-used computer model for predicting the coverages of beacons when planning DGNSS systems. Of comparable importance, however, is ensunng that the beacons' signals are available with an adequate probability across that coverage region. This research analyses this question ofDGNSS radiobeacon availability. It identifies and quantifies the many factors which detennine the probability of obtaining a service of adequate quality. Stochastic elements, such as atmospheric noise and skywave propagation, are evaluated. So are detem1inistic factors such as groundwave propagation of the beacon's signal, and interference. Novel techniques are then proposed for combining these multiple factors so as to allow a single probability of availability to be calculated at any location. This work is then extended to include continuity: the probability of the service remaining available o ver a specified period. These new techniques have been built into a computer model that evaluates the availability, continuity, and coverage for both individual stations and networks of DGNSS radiobeacons. The resulting software enables administrations to plan their systems, ensuring that all three criteria are met.