The Problem of Volcanic Unrest: The Campi Flegrei Case History

Abstract

Campi Flegrei is one of the most densely populated active volcanic areas of the world. It consists of a complex structure with a multicollapse caldera and many postcaldera, explosive, mostly monogenetic vents. The last eruption was in 1538 (Mt. Nuovo). Progressive reduction of eruption energy has accompanied migration of the vents toward the center of the caldera. Campi Flegrei activity is fed by a shallow magma chamber in which trachybasaltic magma undergoes cooling and fractionation, producing trachytic residual melt that is periodically tapped and erupted to the surface. The depth of the chamber is estimated, from petrological and active and passive seismic data, to be about 4 km beneath the town of Pozzuoli. In case of volcanic reactivation, 200 000 people would be at risk and have to be evacuated before its onset. Campi Flegrei caldera has long been the site of slow vertical movements, earthquakes, and thermal activity (hot springs and fumaroles). In the period 1820–1968 the secular trend of ground motion was downward at an average rate of 14 mm/year. In 1968(?)–1972 and in 1982–1984 two episodes of unrest occurred with strong uplift, marked seismicity, increased fumarolic output, and changed thermal-fluid chemistry, well illustrating the complex problems volcanologists face during such volcanic unrest. After both crises, a net uplift remained (150 cm in 1972; 100 cm in 1984), totaling 2.5 m since 1968. Earthquake hypocenters did not exceed 4 km depth, and epicenters were distributed on the inner caldera margins along inward dip** planes. Although neither crisis ended with an eruption, both caused severe civil defense problems and required temporary evacuation of the old center of Pozzuoli. However, in spite of several portending eruption precursors (uplift, seismicity, geochemical anomalies) no alert was issued for an impending eruption, because at the time no clear signs of magma movement toward the surface were recognized. Data processing after the crisis led the majority of geophysicists and geochemists to favor a model in which the unrest was due essentially to heating and thermal expansion of fluid in confined geothermal aquifers, without any direct involvment of the magma chamber. In this paper we present an analysis of data and models showing that overpressure in the Campi Flegrei magma chamber cannot be ruled out and that the two crises could represent two steps in a prolonged process of volcanic reactivation, as probably occurred before the 1538 eruption.