Volcanic eruptions broke out in Rabaul, New Guinea, last year for the first time since 1943. The recent eruptions were minor, compared with some of the area's past volcanic activity, but nevertheless, they underscore the need to study and plan for future eruptions.
by the Rabaul Petrology Group
Heavy falls of volcanic ash in September 1994 from active volcanoes caused extensive damage to much of Rabaul, a town near Papua, New Guinea, with a population of approximately 15,000. The ash cover was up to a meter thick in parts of the area. Roofs fell in, drains were blocked, power lines collapsed, and trees were stripped of branches and leaves.
Explosive eruptions broke out on September 19, 1994, at Tavurvur and Vulcan volcanoes within the Rabaul caldera complex. The resulting ash falls caused massive damage to Rabaul. The eruptions followed a widely publicized earthquake and ground-uplift crisis between September 1983 and July 1985. This "seismo-deformational" event was part of a longer period of caldera unrest that appears to have begun in 1971. The 1983–1985 crisis received widespread scientific attention because of similar events at two other restless calderas at about the same time: in Campi Flegrei near Naples, Italy, and in Long Valley, California.
Understanding the behavior of restless calderas is one of the major challenges of modern volcanology because of uncertainties about the kind of phenomena that precedes different scales of activity. We have little idea how to predict whether a large caldera complex will produce a minor eruption, or a large ignimbrite- and caldera-forming event that would devastate populated areas, or a relatively benign eruption that develops into a catastrophic one. To help resolve these uncertainties we studied the geochemistry and petrology of blocks, pumices, and ashes produced during the 1994 eruptions.
The harbor at Rabaul (Figure 1) is a sea-flooded caldera complex that has grown from a series of collapse events, each probably associated with the major eruption of ignimbrites and related tephras that now form a plateau to the south and west of the harbor area. The average repose period between ignimbrite eruptions over the last 18,000 years is between about 2000 and 3600 years. The most recent of these large eruptions took place about 1400 years ago. There is some evidence for mixed-magma eruptions at Rabaul, and some scientists have suggested that mafic magmas may help mobilize the more felsic magmas of the Rabaul volcano and help them rise to the surface. Earthquakes at Rabaul occur within a 10 x 5 km area of seismicity that appears to represent the ring-fault boundaries of an active caldera block, perhaps formed as a result of the 1400-year-old eruption.
Tavurvur and Vulcan are two eruption centers within the Rabaul caldera complex; they were formed during the Holocene. Virtually simultaneous eruptions at both volcanoes are known to have taken place in 1878, 1937, and in 1994, which makes for a 57–59 year interval between these latest three events. Poorly documented eruptions took place in 1767 and 1791 and possibly involved only Tavurvur.
Fig. 1. Rabaul caldera complex and the intracaldera volcanoes of Tavurvur and Vulcan, which produced eruptions in 1994. The main centers of eruption are represented by asterisks
The 1983–1985 episode of unrest at Rabaul has been described as "perhaps the most threatening example" of unrest events at 139 calderas worldwide. Two Rabaul caldera earthquakes on September 18, 1994, marked the beginning of a seismic swarm that concentrated in the Tavurvur and Vulcan areas. The swarm included many earthquakes, and evacuation of the town began later that day. Increased stages of alert were recommended by the Rabaul Volcanological Observatory (RVO), and most townspeople had left Rabaul by the start of the eruption the next day. The September 19 eruption began at Tavurvur, producing an ash plume that was driven by southeast winds toward the town, and which caused darkness and ashfalls. Vulcan joined Tavurvur in activity 71 minutes later.
The two volcanoes had contrasting eruptions. Vulcan produced a violently explosive column up to about 20 km, while the Tavurvur eruption cloud reached a height of about 6 km. Tavurvur produced mainly dark-brown, fine-grained, phreatomagmatic ash, but Vulcan material was mainly pumice and lighter-colored, coarser ash. Vulcan activity declined markedly on the fifth day of the eruption and had ended by October 2. The eruptions of 1994 resemble those of 1937. In both 1937 and 1994, large caldera earthquakes struck the area just 27 hours after volcanic activity began, leaving very little time for warnings and evacuation.
We selected 21 samples of ejected lava blocks, pumice, and ash from the 1994 Tavurvur and Vulcan eruptions for chemical analysis, and for mineral and glass analysis. Whole-rock samples were analyzed for elements. The rocks are almost entirely high-K, high-SiO2 andesite. The 1994 Rabaul magmas have high-SiO2 chemical compositions similar to those of major ignimbrite-forming eruptions at Rabaul although they are more toward the mafic end of the spectrum.
Unrest at caldera complexes is by itself a justifiable reason for giving attention to major ignimbrite-generating potential, particularly at calderas such as Long Valley where there are no known historical eruptions and at Campi Flegrei, where the latest eruption was in the 16th century. Rabaul has had several eruptions during the past 200 years, so perhaps explosive potential is released rather readily, possibly from a fractured caldera that prevents the build up of substantial magma pressures over centuries or longer. However, a much better understanding of the dynamics and evolution of the type of magma underlying Rabaul is required before a confident prognosis can be made of the scale of eruptions at Rabaul in the future.
Source: Eos, April 25, 1995, p. 171.
Return to Science and
Society
Return to Starting
Point
