Mauna Loa hasn't quite slipped back into a full slumber, but as the new year dawns, the world's largest volcano does appear to be dozing.

The rate of earthquake activity below the mountain, which last erupted in 1984, dropped off dramatically in 2005 from the previous year.

"Last December, most people were predicting an eruption soon. The story 12 months later is completely different," said geophysicist Mike Poland of the Hawaiian Volcano Observatory.

Meanwhile, no end is in sight to Kilauea's Pu'u 'O'o eruption, which began Jan. 3, 1983. The volcano continues to spew out an average of 500,000 cubic yards of lava per day.

Flows from that 23-year-old eruption have destroyed 189 structures and added about 600 acres to the Big Island, but in recent years the lava has shifted west toward earlier flows and into the ocean, drawing thousands of visitors to Hawai'i Volcanoes National Park.

A Mauna Loa eruption also has the potential to devastate surrounding communities, particularly if lava is pushed out from rift zones on the volcano's flanks. The 1984 eruption came within four miles of Hilo, and since then, more than $2 billion of development has occurred on the 13,677-foot-tall mountain, which covers half the island.

"We're kind of worried. We live right by a big rift that's been active before and it can happen again," said Ken Wicks of Hawaiian Ocean View Estates, a rural subdivision in Ka'u at risk from an eruption in Mauna Loa's southwest rift zone.

"It's always in the back of our mind. We never forget it. It is a reality," said Wicks, who is president of the Ocean View Chamber of Commerce.

Wicks welcomed news that Mauna Loa appears to be settling down a bit, and he said lack of rain is of more immediate concern to area residents.

Mauna Loa has erupted 33 times since its first well-documented event in 1843. It began stirring again in mid-2002, at which time scientists detected swelling of the summit. In 2004, unprecedented levels of seismic activity raised concerns that an eruption was imminent. Approximately 2,000 earthquakes were detected underneath Mauna Loa, many of them characterized as deep, long-period events that suggested upward movement of magma.

But the seismic activity backed off at the beginning of 2005. In January, only 34 earthquakes were noted, compared with 365 in December 2004, according to the Hawaiian Volcano Observatory. By the end of 2005, a little more than 300 temblors had been recorded directly under Mauna Loa.

"Relative to last year, I'd say we are further from an eruption," Poland said. "Those thousands of very deep earthquakes more or less shut off at the beginning of the year. We're still getting more than normal but certainly nothing like in 2004."

Inflation at the summit continued through the first half of 2005, but slowly began leveling off. "It is still inflating and we know magma is accumulating, but it appears to be less unstable than at the end of last December," Poland said.

"But it could be just an appearance."

The constant and intense seismic activity in 2004 could have opened bigger pathways, he said, so that magma didn't have to crack through more rock to continue its buildup toward an eruption.

Since scientists can't peer under the mountain to see whether that is actually happening, much was accomplished last year to expand detection of changes on the surface.

The observatory, in cooperation with the University of Hawai'i -Manoa, installed 11 new continuously recording global positioning satellite instruments on Mauna Loa, almost doubling the number of GPS stations on the volcano. Most of the new sites are along the northeast and southwest rift zones, where magma is most likely to emerge in a non-summit eruption.

"We're always looking for ways to determine when magma is going to go out into a rift zone," said Jim Kauahikaua, the observatory's scientist-in-charge, since eruptions on the volcano's slopes can reach homes more quickly than a summit event.

Scientists also installed instruments at the summit caldera to continuously record vent temperatures and sulfur dioxide and carbon dioxide concentrations from vapors emitted as a result of subsurface activity. The measurements are transmitted to the observatory every 10 minutes. Previously, these measurements were obtained from helicopter flyovers every few months.

A panoramic camera placed at the Mauna Loa summit in 2005 provides images every few minutes that can be used to verify reports of eruptions and other substantial changes at the caldera. "It's a pair of eyes always up there and always looking," Poland said. The photos are not yet available online to the public.

Poland added his expertise in radar interferometry to the Mauna Loa monitoring mix when he joined the Big Island observatory in 2005 from the Cascades Volcano Observatory in Washington. The technique, known as InSAR, detects changes in the shape of the ground surface by comparing radar images taken from a satellite.

Poland said InSAR can provide a "whole picture" of deformation patterns when used with GPS data gathered from specific points on the volcano.

Despite scientific advances and improved monitoring of Mauna Loa, Kauahikaua said scientists still are unable to answer the question most on the minds of Big Island residents: When is the volcano going to erupt?

While observatory staff are confident the mountain will tell them when an outbreak of lava is hours or mere days away, predicting an eruption months or years out is not yet possible.

Poland said, "One of the things that's perpetual is that as we continue to put more and more instruments out there, the more we learn about the volcano, the more questions are raised."

Scientists studying Kilauea are looking for answers to their own questions, including why this event has continued for more than two decades, when the volcano's 10 previous eruptions lasted no longer than a month.

The Pu'u 'O'o eruption is the longest rift-zone eruption in at least 600 years. Geologist Christina Heliker said the cone, 12 miles downhill from the main caldera, apparently has an open tap into the magma reservoir underneath Kilauea, and that "there's plenty more to go."

While calderas usually have a direct connection to deep magma zones, she said, cones and vents shooting off from the main line can be expected to close off at some point. Not so with Pu'u 'O'o.

"All the pathways leading to the surface are just unusually efficient in this eruption and keep the thing going," Heliker said. "The only thing that could stop the flow is a large flank earthquake."

The most dramatic developments in 2005 in the Kilauea eruption occurred at the East Lae'apuki ocean entry, which began in early May. As the resulting lava bench grew rapidly, small collapses occurred, prompting the National Park Service to close the area to visitors in July.

By early August, the bench had grown to 30 acres. On Aug. 27, a collapse claimed 11 acres of the newly created coastline.

The lost acreage was replaced with 34 acres of fresh lava, leading to a series of collapses Nov. 28 that took the entire bench plus 10 acres of adjoining terrain. Steam explosions from exposed lava tubes mixing with sea water propelled molten spatter and rock 100 yards inland.

The observatory reports it was the largest ocean-entry collapse in the history of this Pu'u 'O'o eruption.

The process continues, with a new bench forming.

Reach Christie Wilson at cwilson@honoluluadvertiser.com.