Hawaiian Volcano Observatory
yellow horizontal separator line

skip past main content navigational bar Kīlauea

yellow horizontal separator line

Mauna Loa

yellow horizontal separator line


yellow horizontal separator line

Other Volcanoes

yellow horizontal separator line

Volcanic Hazards

yellow horizontal separator line

About HVO

yellow horizontal separator line


April 8, 2010

A weekly feature provided by scientists at the Hawaiian Volcano Observatory.

What caused Kīlauea to lose its top in about 1500?

All of us lose something as we age, and, about 500 years ago, Kīlauea lost its top. The summit collapsed to form a precipitous hole, known as a caldera, about 3 km (2 miles) across and probably more than 600 m (2,000 feet) deep.

The life-changing event likely occurred over a few days or weeks and is arguably recorded by oral tradition describing Hi`iaka's digging for the body of Lohi`au, whom Pele had killed in a love triangle gone sour.

The caldera has been filling with lava flows since 1790 and now reveals only a shadow of its former, much deeper self.

What caused the caldera to collapse? We don't know, but we do know quite a bit that helps us focus on the possible causes.

To trigger collapse, material must be removed from the magma reservoir within the volcano, creating a void into which the overlying rocks fall. The generally accepted notion is that magma is rapidly erupted, emptying the reservoir, and the volume of the ensuing collapse roughly equals the volume of magma erupted.

We can estimate the volume of Kīlauea's caldera as 4-6 cubic kilometers (1-1.4 cubic miles). A lava flow of such volume that erupted rapidly about 500 years ago could be the cause of the collapse.

But no such lava flow exists, either on the island or on the Puna Ridge, the submarine extension of the east rift zone. Hmmm. Now we start scratching our heads to think of a reason for the summit collapse.

One good possibility is that magma drained the reservoir quickly but stayed underground rather than erupting. Such a possibility would require the intrusion of one or more enormous dikes. For example, a dike 40-60 km (25-35 miles) long, 10 km (6 miles) high, and 10 m (30 feet) wide would contain a volume of magma equal to that of the caldera.

Observed or inferred dikes in Hawai`i are far smaller-a few kilometers (miles) long at most, 3 km (2 miles) high or less, and 1-2 m (3-6 feet) wide. Nonetheless, the injection of a huge dike or family of dikes cannot be discounted and may eventually be acknowledged as the cause of collapse.

Any eruption or intrusion draining a magma reservoir and causing caldera collapse is thought to take place rapidly. The reason is that large voids should not remain open for long under the crushing weight of rocks above. But what if this reasoning is wrong? Could a large but long-lasting eruption eventually drain a magma reservoir, slowly creating void space that eventually pancakes out by collapse?

We are led to such an outrageous hypothesis by the observation that, over a 60-year period just before caldera collapse, a lava flow was erupted from Kīlauea with a volume equivalent to that of the caldera. That lava flow, called the `Aila`au flow, erupted from the Thurston lava tube area and eventually covered most of Kīlauea north of the east rift zone. This is the largest lava flow known on Kīlauea, and it was followed immediately by collapse of the caldera. Linking such two large events is satisfying, but satisfaction doesn't make the outrageous hypothesis correct.

In fact, there are at least three major problems with it. First, could the summit hold itself up as voids were growing below it? Most geologists would scoff at this. But if we view the magma reservoir not as a balloon, but instead as a series of interconnected bodies of magma with supports between, akin to a mine, perhaps void space becomes more palatable.

If slow eruption were emptying the reservoir, no magma would be rising from below to replenish it. This may be a major problem for geophysicists who consider the supply of magma from the hot spot to be continuous.

Finally, if the reservoir were emptying (depressurizing) during the `Aila`au eruption, where was the pressure coming from that causes the eruption to continue?

The outrageous hypothesis will be presented at an international conference about calderas this fall at Piton Fournaise, a near-twin of Kīlauea on Reunion Island in the Indian Ocean. Perhaps "outrageous" will be changed to "impossible" or maybe even "possible" after this conference.

Kīlauea Activity Update

On Kīlauea's east rift zone, breakouts along the east margin of the Thanksgiving Eve Breakout (TEB) flow remain active on the Pulama pali. The active flow front was at the 700 ft elevation level on Thursday, April 8. If lava supply to these flows continues, they are expected to eventually rejoin the main TEB flow field just west of the County Viewing Area. The flows may also stall before reaching the coastal plain due to ongoing lava supply fluctuations.

At Kīlauea's summit, a ponded, circulating lava surface deep within the collapse pit inset within the floor of Halema`uma`u Crater was visible via Webcam during much of the past week. Volcanic gas emissions remain elevated, resulting in high concentrations of sulfur dioxide downwind.

One earthquake beneath Hawai`i Island was reported felt during the past week. A magnitude-1.3 microearthquake occurred at 9:22 a.m., H.s.t., on Friday, April 2, 2010, and was located 2 km (1 mile) northwest of `Opihikao, at a depth of 2 km (1 mile).

Visit our Web site ( for detailed Kīlauea and Mauna Loa activity updates, recent volcano photos, recent earthquakes, and more; call (808) 967-8862 for a Kīlauea activity summary; email questions to

skip past bottom navigational bar

Homeblank spacerVolcano Watchblank spacerProductsblank spacerGalleryblank spacerPress Releases
How Hawaiian Volcanoes Work

The URL of this page is
Updated: April 20, 2010 (pnf)