San Francisco — Hawaii's recent double eruption of Mauna Loa and Kilauea - the first since 1868 - is a ''scientific bonanza.'' That is the conviction of Robert Decker, scientist in charge of the United States Geological Survey's Hawaiian Volcano Observatory. He took a few minutes out of his hectic schedule recently to discuss the studies that he and his 10 colleagues are pursuing as the two volcanoes put on their spectacular fireworks display.
The Mauna Loa eruption came as no surprise. Last fall, at a meeting of the American Geophysical Union, Dr. Decker and his colleagues predicted such an event within two years.
''There has been a steady evolution in our predictive ability,'' he explained in a phone interview.
There are two primary tools that the scientists use to forecast the volcanoes' behavior. One is the precise measurement of seismic activity deep within the mountain. As much as a year ahead of time the pattern of small swarms of ''microquakes'' intensifies.
The second is increased swelling of the volcano itself. The scientists put a series of reflectors on the caldera. Using very precise laser surveying equipment, they can measure very small changes in the reflectors' positions. In this way, they have discovered that Mauna Loa swells at a faster rate before an eruption.
''Although improved, our predictive capability is not precise, not by any means,'' explains the scientist. While increased seismic activity and an acceleration in the rate of deformation are definite precursors, they are only long term signs.
''I was very disappointed that there was no good, medium-term warning: A week before, we had no reason to suspect that this week there would be an eruption,'' Decker explains.
The mountain, however, did give a definite warning an hour before it went off. An hour and 15 minutes before the outbreak, there was an intensification in the earthquake swarms coming from it. Then, just 45 minutes before a tremor, a continuous vibration, which indicates magma motion, began.
The current activity is providing the scientists with a lot of good data, Decker says. The last eruption in 1975 was very quick, so they didn't have much time to study it. Today the 10 scientists at the observatory are better organized. ''This will give us the good baseline data that we need. As a result, we should be able to do a much better job of prediction next time,'' the scientist says.
The scientists have already determined that the chamber which holds the molten rock is much shallower than they had thought. And the simultaneous eruption of Kilauea has allowed them to put to rest another longstanding question: There is no direct connection between the magma chambers of the two volcanoes although they are only 25 miles apart.
Unfortunately for those living near these volcanoes, the ability to predict the course that flowing lava will take lags that of predicting when eruptions will occur. Generally, like water, lava will flow down the steepest slope. Unlike water, however, it has a tendency to put out lobes of material to one side or the other. When it does this, the lobe can deflect the lava stream by as much as 10 degrees, even more on flatter land.
''There seems to be quite an element of randomness in these zigs and zags. As a result, we've decided it is not productive to speculate about the lava's course: We end up alarming some people more than they should be and giving other people a false sense of security,'' Decker explains.
At press time yesterday, Hawaii civil defense officials said the leading edge of the lava flow was within four miles of the outskirts of Hilo, 12 miles from the heart of the city. But officials assured residents no area of Hilo was in immediate danger and residents would be given 24 hours notice to evacuate.