Mr. MCKELVEY. Yes; the picture on the upper left is Van Norman Dam, and another section-

Senator BIBLE. Did you predict that in advance of the actual earthquake?

Mr. MCKELVEY. We are unable to predict the time at which an earthquake will occur. But we believe that if this program is allowed to go forward, we will be able to predict some kinds of earthquakes, including those along the San Andreas Fault. We will be able to develop that capability within a period of several years. A good deal of progress has been made in understanding faults of the San Andreas type, and we are gaining confidence in the probability of developing a prediction capability.

If that capability is developed, then there are many steps that can be taken. People can be evacuated, and safety equipment can be protected. Much of the damage of earthquakes, of course, comes not just from the shaking itself, but from the secondary phenomena such as fires that take place afterward. Gas can be cut off. Steps of that sort can be taken that can be highly effective.

MOVEMENT ALONG SAN ANDREAS FAULT

Senator BIBLE. Are you convinced that there will be future earthquakes along the San Andreas Fault?

The Pacific plate moves northwestward relative to the American plate at a rate of about 1.5 inches per year, but movement of brittle, near surface rocks cut by the San Andreas fault, is discontinuous. Segments of the fault are creeping and produce movements about equal to the plate movements; other segments of the fault are locked and are temporarily storing elastic strain energy. Large earthquakes can occur along any part the the fault system, but the likelihood of a great earthquake is higher along the temporarily locked segments. Mr. MCKELVEY. Without question. The San Andreas Fault is what is known geologically as a transform fault. It is a contact between two major plates of the earth is crust that are moving steadily. The movement along those plates amounts to about an inch and a half a year.

The San Andreas does not move uniformly. Parts are creeping steadily. Shown on this map, for example, are areas in which the fault is creeping steadily at the rate of about an inch and a half a year. Also shown are areas in which the fault is locked, where no movement is taking place.

Senator BIBLE. No movement?

In areas of fault creep, gradual displacements offset curbstones and cause relatively minor damage to the works of man. The graph below the left-hand picture shows that movement of the fault essentially keeps pace with the relative movement of the plates. As shown at the right, in areas where active faults are temporarily locked, large amounts of elastic strain are accumulated, and large, damaging earthquakes can occur.

Mr. MCKELVEY. Yes; so the strain is gradually accumulating, and when that seismic strain reaches the point of rupture then there is a catastrophic release, which amounts to a very severe earthquake. We have shown on this chart the difference between these two kinds of phenomena. Here in the Hollister area of California on the left, we have shown the curve in a curbstone. The fault is creeping gradually, moving just as it would in a catastrophic movement, but its effects are at a minimum because it is moving slowly and gradually.

The San Andreas Fault has been moving for about 60 million years, and, I believe, has a total displacement of about 300 miles.

Senator BIBLE. What is the depth and height or whatever you call it of the fault.

Mr. MCKELVEY. The movement along the San Andreas is primarily horizontal.

Senator BIBLE. The fault where I grew up had a vertical movement, and some of them were a couple of feet, or something like that.

Mr. MCKELVEY. Many of those in Nevada have primarily a vertical component. Some have both a vertical and horizontal component. Senator BIBLE. When was the San Francisco earthquake, 1906 ? Mr. MCKELVEY. Yes.

Senator BIBLE. Was the movement vertical?

Mr. MCKELVEY. Primarily horizontal, and I believe it had displacements of up to more than 20 feet in places. But the total displacement along the fault over time, as I said, has been about 300 miles, and that movement is continuing. To come back to your original question, will there be additional earthquakes along that fault? The answer is unquestionably yes. That movement is continuing, and unless ways can be found to have this movement take place gradually over the entire fault, there will be severe earthquakes.

SEISMIC EFFECT OF UNDERGROUND NUCEAR EXPLOSIONS

Senator BIBLE. One further question on the earthquake problem, and I will turn the earthquake to our Senators from Alaska and Oregon, if they have any questions.

What effect do underground nuclear energy explosions, and we have had many of them in Nevada, and a big one in central Nevada, what effect do they have on the creation of earthquakes? Do they cause earthquakes?

Mr. MCKELVEY. They cause ground shaking in the immediate vicinity of the explosion in an amount that is related to the energy that has been released. In some areas, I believe, they have also caused some displacement. Some of the adjustment that has taken place in the ground in the vicinity of the explosion, has taken place on preexisting faults; there has been some slight fault movement. These have only been minor displacements, but there has been some slight movement along preexisting faults in some instances.

Senator BIBLE. Senator Stevens, do you have any earthquake questions?

MEASUREMENT OF TENSION

Senator STEVENS. Yes, I do. Have one related to your comment about the static areas. Is it possible to measure the tension that is building up in these areas?

Five seismically active areas in the Western conterminous United States will be the focal points of earthquake prediction research. The diagram on the right shows a stylized representation of a dense monitoring network along the San Andreas fault, where most initial efforts toward earthquake prediction will be concentrated

Mr. MCKELVEY. This is possible, we think, and this is one of the things we are attempting to do with our observation net that we are developing along the San Andreas fault.