We have a large amount of data from the last 7 years of the Montserrat eruption. For the analysis of seismic parameters on such a large dataset, an automated event picking and classification program is the most practical approach.

The AECAP searches through the continuous data files, finding the events, using a long term average (LTA) - short term average (STA) method to trigger events. This method does not work for tremor. If there is no event triggered during a specified time period, a tremor subroutine looks for high amplitude tremor.

Subroutines classify them and calculate useful seismic parameters for analysis.

The classification subroutines work by using the frequency content, the impulsiveness of the event onset and the amplitude of the events.

The program output contains the following parameters:

The event classification

Start time of the event

dt - time between the start of 2 events, both between like events and previous events

Duration of event

Peak amplitude

Relative energy

Peak frequency

How much of the energy is below 2 Hz.