If you are choosing a lightning safety system, one decision shapes everything else: should alerts be based on predicted risk, on reported lightning detections, or on both? The two approaches sound similar but work differently, and understanding the difference helps you pick the one that fits your site, your policy, and your response process.
Prediction-based systems, sometimes called electrostatic systems, measure changes in the atmospheric electric field above a site and issue an alert when the field crosses the risk threshold the system is configured to watch for. The aim is to provide notice before a nearby strike. The National Severe Storms Laboratory notes that forecasting exactly where and when an individual strike will occur is not possible, so a prediction is an estimate of risk rather than confirmation of a strike.
Detection-based systems take a different input. Instead of estimating from a local field, they use reports of lightning that has actually been observed and measure each reported strike against the proximity thresholds you set. How lightning detection works covers the mechanics, but the short version is that detection reports observed strikes and their distance from your site rather than estimating risk.
The practical question is which input your team wants to act on: an estimate of atmospheric conditions, or a report of observed strikes measured against your configured policy. Weatherstem is built on detection. The Practical Lightning Assistant lets you draw an alert ring around the area you need to protect, sends alerts when lightning is reported inside it, and issues an automatic all clear once your configured interval passes with no new reported strikes.
A detection system depends on the data feeding it, so the source matters. Weatherstem uses the National Lightning Detection Network as its primary lightning-data source, with AccuWeather as a secondary source intended to support continuity. When you evaluate any system, it is worth asking what data it relies on and what happens if that source is disrupted.
Whichever approach you choose, the value comes from a documented, repeatable process. With detection-based alerting, every reported strike, alert, and all clear is timestamped and recorded, so you can review an event afterward. For how to turn that into a written plan, see our guide to a defensible lightning safety policy.
Prediction estimates lightning risk from changes in the local atmospheric electric field. Detection uses reports of observed strikes measured against the thresholds you configure. One estimates conditions, the other reports observed activity.
Lightning can be reported well beyond the rain, so a detection system watching a wide ring around your site can alert you before a storm is overhead. You set the distance threshold that fits your operation.
If you are reviewing your current setup, see how detection-based alerting compares on our lightning system replacement page, or book a short demo.