The likelihood of increased geomagnetic activity can be estimated by examining the various daily solar observations to determine whether CMEs have occurred or whether coronal holes have been observed. ACE and SOHO solar wind data as well as other satellite and ground based measurements of the Sun also help determine whether any solar disturbances are definitely Earth-bound, or whether the data may even hint at imminent arrival at Earth.
In making an assessment of geomagnetic hazard, thought needs to be given to the start time, the duration and the magnitude of geomagnetic activity. We know that most solar disturbances will take 1-4 days to travel to Earth orbit, but the accurate assessment of geomagnetic storm onset and development is still a science in its infancy. In the future we can anticipate increasingly sophisticated models of the Sun, the solar wind and the Earth's magnetic environment. But there are already many forecast tools and public domain resources for predicting geomagnetic hazard, based on observations, in use around the world.
We have our own tools at BGS. Some of our data and products are shown here:
Finally, geomagnetic hazard forecasts usually need to be tailored for particular customers, as they don't necessarily understand (or want to understand!) the jargon of the space or geomagnetic scientist. For instance, satellite operators, power grid engineers and pipeline operators often want different things and forecaster experience and ability to appropriately interpret models and data is vital.