Prediction of Volcanic Activity - General Principles

General Principles

Various methods including the following sections are used to help predict eruptions. In using these methods, five major principles form the basis of eruption forecasting is as follows:

• the principle of inflection points in trends states that with unknown rates of change, a point in time is reached at which the volcanic system becomes unstable and likely will erupt;
• the principle of coinciding change states that one monitored parameter alone may not yield significant symptoms to diagnose an imminent eruption, but unrelated trends of several monitored parameters may start co-evolving as the system approaches a state of instability;
• the principle of known behavior treats a volcano as if it were a medical patient, assuming that responses to changes in the underground may be highly individual to a volcano's particular internal structure and can become better known by understanding its past eruptive characteristics;
• the principle of unexpected behavior treats volcanoes, the public, and decision-makers alike as inherently inconsistent systems - leading to unexpected eruptions (e.g., fast magma ascent from unexpected depth), and mitigation failures;
• the principle of symptom-based short-term forecast as with all the other principles is similar to an epidemiological diagnosis, whereby forecasts are based on symptoms and patient history.

Volcanic eruptions can to date not be predicted by stochastic methods, but only by catching early symptoms before an imminent eruption. Therefore, continuous monitoring even of dormant volcanoes, though costly, is the only way to enable eruptive behavior forecasts. The following sections describe individual groups of methods typically deployed in monitoring volcanoes and the symptomatic evolution of their activity.