Abstract
Small arrays of borehole strainmeters, together with other deformation measurements, allow much tighter constraints on defining the underground movement of magma before and during eruptive activity; higher resolution of these changes over short time scales is possible because of the much higher sensitivity of strain measurements compared with displacement monitoring. We present data recorded during activity at Montserrat in the Caribbean and at Hekla in Iceland. For Montserrat we recorded high signal to noise ratio data during a small Volcanian explosion in March 2004 and show that the data require both a deep reservoir (at about 5 km depth) and a shallow dike. Importantly, we see clear strain changes during about a minute before the surface breakout as magmatic material moves from the reservoir into the expanding dike. By applying conservation rules we are able to determine the bulk modulus of the material in the reservoir. This is about 7 GPa and corresponds to a few percent of free gas, a result in accord with other estimates. The 2000 eruption of Hekla was much better monitored than the previous activity (including more extensive campaign GPS and InSAR) and allow us to determine with considerable confidence the magmatic plumbing of the volcano. The deep reservoir is at about 11 km depth but fissuring at the surface is due to a dike that does not penetrate more than 1 km deep. We propose that the connection between the reservoir and the shallow dike is via a conduit that now remains fluid during the relatively short 10 years between eruptions, and that this is why the eruption interval changed abruptly from about 80 years to 10 years following the large 1947 eruption. As with Montserrat, we record strain changes before the surface breakout and can determine the bulk modulus for the reservoir; in this case it is extremely high (>40 GPa) implying no free gas in the reservoir.
Alan Linde is a geophysicist at the Department of Terrestrial Magnetism of the Carnegie Institution of Washington. For more years than he cares to admit, he has been using borehole strain data to help understand the physics of earthquake failure and of volcanic activity. Together with colleague Selwyn Sacks, he has worked to install small arrays of strainmeters in a number of tectonically active areas with
the objective of capturing seismic and volcanic activity. By collaborating with researchers in other countries they have had success in areas such as California, Iceland, Italy, Greece and Taiwan.